diff --git a/assets/models-dev/catalog.json b/assets/models-dev/catalog.json
index ea5fd04624..92aee15a1b 100644
--- a/assets/models-dev/catalog.json
+++ b/assets/models-dev/catalog.json
@@ -6937,6 +6937,56 @@
         "temperature": true,
         "tool_call": true
       },
+      "claude-fable-5": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 1.1,
+          "cache_write": 13.75,
+          "input": 11,
+          "output": 55
+        },
+        "description": "Claude model for creative writing, analysis, and controlled agent workflows",
+        "family": "claude-fable",
+        "id": "claude-fable-5",
+        "interleaved": true,
+        "knowledge": "2026-01-31",
+        "last_updated": "2026-06-09",
+        "limit": {
+          "context": 1000000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Claude Fable 5",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "low",
+              "medium",
+              "high",
+              "xhigh",
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-06-09",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
       "claude-opus-4-6": {
         "attachment": true,
         "cost": {
@@ -7469,6 +7519,56 @@
         "temperature": true,
         "tool_call": true
       },
+      "claude-sonnet-5": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.2,
+          "cache_write": 2.5,
+          "input": 2,
+          "output": 10
+        },
+        "description": "Everyday Claude agent model for coding, planning, browsing, and general work",
+        "family": "claude-sonnet",
+        "id": "claude-sonnet-5",
+        "interleaved": true,
+        "knowledge": "2026-01-31",
+        "last_updated": "2026-06-30",
+        "limit": {
+          "context": 1000000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Claude Sonnet 5",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "low",
+              "medium",
+              "high",
+              "xhigh",
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-06-30",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
       "coding-glm-5.1": {
         "attachment": false,
         "cost": {
@@ -8479,6 +8579,52 @@
         "temperature": true,
         "tool_call": true
       },
+      "gemini-3.5-flash": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 1.5,
+          "input": 1.5,
+          "output": 9
+        },
+        "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
+        "family": "gemini-flash",
+        "id": "gemini-3.5-flash",
+        "knowledge": "2025-01",
+        "last_updated": "2026-05-19",
+        "limit": {
+          "context": 1000000,
+          "output": 64000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "audio",
+            "video"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Gemini 3.5 Flash",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "minimal",
+              "low",
+              "medium",
+              "high"
+            ]
+          }
+        ],
+        "release_date": "2026-05-19",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
+      },
       "glm-5.2": {
         "attachment": false,
         "cost": {
@@ -9054,6 +9200,150 @@
         "temperature": false,
         "tool_call": true
       },
+      "gpt-5.6-luna": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.1,
+          "cache_write": 1.25,
+          "input": 1,
+          "output": 6
+        },
+        "description": "Cost-efficient GPT-5.6 model for fast, high-volume workloads",
+        "family": "gpt-luna",
+        "id": "gpt-5.6-luna",
+        "knowledge": "2026-02-16",
+        "last_updated": "2026-07-09",
+        "limit": {
+          "context": 1050000,
+          "input": 922000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "GPT-5.6 Luna",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "none",
+              "low",
+              "medium",
+              "high",
+              "xhigh",
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-07-09",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
+      "gpt-5.6-sol": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.5,
+          "cache_write": 6.25,
+          "input": 5,
+          "output": 30
+        },
+        "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows",
+        "family": "gpt-sol",
+        "id": "gpt-5.6-sol",
+        "knowledge": "2026-02-16",
+        "last_updated": "2026-07-09",
+        "limit": {
+          "context": 1050000,
+          "input": 922000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "GPT-5.6 Sol",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "none",
+              "low",
+              "medium",
+              "high",
+              "xhigh",
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-07-09",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
+      "gpt-5.6-terra": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.25,
+          "cache_write": 3.125,
+          "input": 2.5,
+          "output": 15
+        },
+        "description": "Balanced GPT-5.6 model for capable, cost-efficient everyday work",
+        "family": "gpt-terra",
+        "id": "gpt-5.6-terra",
+        "knowledge": "2026-02-16",
+        "last_updated": "2026-07-09",
+        "limit": {
+          "context": 1050000,
+          "input": 922000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "GPT-5.6 Terra",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "none",
+              "low",
+              "medium",
+              "high",
+              "xhigh",
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-07-09",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
       "grok-4.3": {
         "attachment": true,
         "cost": {
@@ -9113,6 +9403,123 @@
         "temperature": true,
         "tool_call": true
       },
+      "grok-4.5": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.5,
+          "input": 2,
+          "output": 6
+        },
+        "description": "xAI's latest Grok for chat, coding, agentic tools, and lower hallucination risk",
+        "family": "grok",
+        "id": "grok-4.5",
+        "last_updated": "2026-07-08",
+        "limit": {
+          "context": 1000000,
+          "output": 1000000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Grok 4.5",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "none",
+              "low",
+              "medium",
+              "high"
+            ]
+          }
+        ],
+        "release_date": "2026-07-08",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
+      },
+      "grok-build-0.1": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.2,
+          "input": 1,
+          "output": 2
+        },
+        "description": "Fast Grok coding model tuned for agentic engineering and iterative edits",
+        "family": "grok-build",
+        "id": "grok-build-0.1",
+        "last_updated": "2026-04-16",
+        "limit": {
+          "context": 256000,
+          "output": 256000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Grok Build 0.1",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [],
+        "release_date": "2026-04-16",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
+      },
+      "hy3-preview": {
+        "attachment": false,
+        "cost": {
+          "cache_read": 0.051,
+          "input": 0.17,
+          "output": 0.566661
+        },
+        "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
+        "family": "Hy",
+        "id": "hy3-preview",
+        "last_updated": "2026-04-20",
+        "limit": {
+          "context": 256000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Hy3 Preview",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "none",
+              "low",
+              "high"
+            ]
+          }
+        ],
+        "release_date": "2026-04-20",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
+      },
       "kimi-k2.5": {
         "attachment": true,
         "cost": {
@@ -9197,6 +9604,90 @@
         "temperature": false,
         "tool_call": true
       },
+      "kimi-k2.7-code": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.160835,
+          "input": 0.95,
+          "output": 3.9995
+        },
+        "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
+        "family": "kimi-k2",
+        "id": "kimi-k2.7-code",
+        "interleaved": {
+          "field": "reasoning_content"
+        },
+        "knowledge": "2025-01",
+        "last_updated": "2026-06-12",
+        "limit": {
+          "context": 262144,
+          "output": 32768
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "video"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi K2.7 Code",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          }
+        ],
+        "release_date": "2026-06-12",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
+      "kimi-k2.7-code-highspeed": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.32167,
+          "input": 1.9,
+          "output": 7.999
+        },
+        "description": "Lower-latency Kimi Code variant for interactive edits and coding-agent loops",
+        "family": "kimi-k2",
+        "id": "kimi-k2.7-code-highspeed",
+        "interleaved": {
+          "field": "reasoning_content"
+        },
+        "knowledge": "2025-01",
+        "last_updated": "2026-06-12",
+        "limit": {
+          "context": 262144,
+          "output": 32768
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "video"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi K2.7 Code Highspeed",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          }
+        ],
+        "release_date": "2026-06-12",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
       "minimax-m2.7": {
         "attachment": false,
         "cost": {
@@ -21280,10 +21771,10 @@
       "moonshotai/kimi-k2.7-code": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.149,
+          "cache_read": 0.16,
           "cache_write": 0,
-          "input": 0.719,
-          "output": 3.49
+          "input": 0.75,
+          "output": 3.5
         },
         "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
         "family": "kimi-k2",
@@ -21316,10 +21807,10 @@
       "stepfun/step-3.7-flash": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.04,
+          "cache_read": 0.03,
           "cache_write": 0,
-          "input": 0.2,
-          "output": 1.15
+          "input": 0.19,
+          "output": 1.14
         },
         "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use",
         "id": "stepfun/step-3.7-flash",
@@ -32648,8 +33139,8 @@
         },
         "last_updated": "2026-06-13",
         "limit": {
-          "context": 1048576,
-          "output": 1048576
+          "context": 256000,
+          "output": 256000
         },
         "modalities": {
           "input": [
@@ -53150,6 +53641,49 @@
         "reasoning": false,
         "release_date": "2024-10-01",
         "tool_call": false
+      },
+      "zai-org/GLM-5.2": {
+        "attachment": false,
+        "cost": {
+          "input": 1.4375,
+          "output": 5.75
+        },
+        "description": "Open flagship GLM for long-horizon coding agents and million-token context work",
+        "family": "glm",
+        "id": "zai-org/GLM-5.2",
+        "last_updated": "2026-06-13",
+        "limit": {
+          "context": 1048576,
+          "output": 131072
+        },
+        "modalities": {
+          "input": [
+            "text"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "GLM-5.2",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "none",
+              "high",
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-06-13",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
       }
     },
     "name": "evroc",
@@ -62529,16 +63063,16 @@
       "gemini-flash-latest": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.075,
-          "input": 0.3,
-          "input_audio": 1,
-          "output": 2.5
+          "cache_read": 0.15,
+          "input": 1.5,
+          "input_audio": 1.5,
+          "output": 9
         },
         "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
         "family": "gemini-flash",
         "id": "gemini-flash-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-19",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -62547,8 +63081,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -62560,15 +63094,16 @@
         "reasoning": true,
         "reasoning_options": [
           {
-            "type": "toggle"
-          },
-          {
-            "max": 24576,
-            "min": 0,
-            "type": "budget_tokens"
+            "type": "effort",
+            "values": [
+              "minimal",
+              "low",
+              "medium",
+              "high"
+            ]
           }
         ],
-        "release_date": "2025-09-25",
+        "release_date": "2026-05-19",
         "structured_output": true,
         "temperature": true,
         "tool_call": true
@@ -62577,14 +63112,15 @@
         "attachment": true,
         "cost": {
           "cache_read": 0.025,
-          "input": 0.1,
-          "output": 0.4
+          "input": 0.25,
+          "input_audio": 0.5,
+          "output": 1.5
         },
         "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
         "family": "gemini-flash-lite",
         "id": "gemini-flash-lite-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-07",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -62593,8 +63129,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -62606,15 +63142,16 @@
         "reasoning": true,
         "reasoning_options": [
           {
-            "type": "toggle"
-          },
-          {
-            "max": 24576,
-            "min": 512,
-            "type": "budget_tokens"
+            "type": "effort",
+            "values": [
+              "minimal",
+              "low",
+              "medium",
+              "high"
+            ]
           }
         ],
-        "release_date": "2025-09-25",
+        "release_date": "2026-05-07",
         "structured_output": true,
         "temperature": true,
         "tool_call": true
@@ -64102,16 +64639,16 @@
       "gemini-flash-latest": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.075,
-          "cache_write": 0.383,
-          "input": 0.3,
-          "output": 2.5
+          "cache_read": 0.15,
+          "input": 1.5,
+          "input_audio": 1.5,
+          "output": 9
         },
         "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
         "family": "gemini-flash",
         "id": "gemini-flash-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-19",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -64120,8 +64657,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -64142,7 +64679,8 @@
             ]
           }
         ],
-        "release_date": "2025-09-25",
+        "release_date": "2026-05-19",
+        "structured_output": true,
         "temperature": true,
         "tool_call": true
       },
@@ -64150,14 +64688,15 @@
         "attachment": true,
         "cost": {
           "cache_read": 0.025,
-          "input": 0.1,
-          "output": 0.4
+          "input": 0.25,
+          "input_audio": 0.5,
+          "output": 1.5
         },
         "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
         "family": "gemini-flash-lite",
         "id": "gemini-flash-lite-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-07",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -64166,8 +64705,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -64188,7 +64727,7 @@
             ]
           }
         ],
-        "release_date": "2025-09-25",
+        "release_date": "2026-05-07",
         "temperature": true,
         "tool_call": true
       },
@@ -86774,9 +87313,13 @@
               "high",
               "max"
             ]
+          },
+          {
+            "min": 1024,
+            "type": "budget_tokens"
           }
         ],
-        "release_date": "2026-04-27",
+        "release_date": "2025-10-15",
         "temperature": true,
         "tool_call": true
       },
@@ -86837,6 +87380,7 @@
         },
         "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control",
         "id": "~anthropic/claude-sonnet-latest",
+        "knowledge": "2025-08-31",
         "last_updated": "2026-05-01",
         "limit": {
           "context": 1000000,
@@ -86867,23 +87411,27 @@
               "high",
               "max"
             ]
+          },
+          {
+            "min": 1024,
+            "type": "budget_tokens"
           }
         ],
-        "release_date": "2026-04-27",
+        "release_date": "2026-02-17",
         "temperature": true,
         "tool_call": true
       },
       "~google/gemini-flash-latest": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.05,
+          "cache_read": 0.15,
           "cache_write": 0.08333333333333334,
-          "input": 0.5,
-          "output": 3
+          "input": 1.5,
+          "output": 9
         },
         "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
         "id": "~google/gemini-flash-latest",
-        "last_updated": "2026-05-01",
+        "last_updated": "2026-05-19",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -86914,7 +87462,7 @@
             ]
           }
         ],
-        "release_date": "2026-04-27",
+        "release_date": "2026-05-19",
         "temperature": true,
         "tool_call": true
       },
@@ -87025,12 +87573,10 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
+              "none",
               "low",
               "medium",
               "high",
@@ -87038,7 +87584,7 @@
             ]
           }
         ],
-        "release_date": "2026-04-27",
+        "release_date": "2026-04-24",
         "temperature": false,
         "tool_call": true
       },
@@ -87070,12 +87616,10 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
+              "none",
               "low",
               "medium",
               "high",
@@ -87083,7 +87627,7 @@
             ]
           }
         ],
-        "release_date": "2026-04-27",
+        "release_date": "2026-03-17",
         "temperature": false,
         "tool_call": true
       }
@@ -87215,6 +87759,87 @@
         "temperature": false,
         "tool_call": true
       },
+      "k3": {
+        "attachment": false,
+        "cost": {
+          "cache_read": 0,
+          "cache_write": 0,
+          "input": 0,
+          "output": 0
+        },
+        "description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
+        "family": "kimi-k3",
+        "id": "k3",
+        "last_updated": "2026-07-16",
+        "limit": {
+          "context": 1048576,
+          "output": 131072
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "video"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi K3",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-07-16",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
+      },
+      "kimi-for-coding-highspeed": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0,
+          "cache_write": 0,
+          "input": 0,
+          "output": 0
+        },
+        "description": "Lower-latency Kimi Code variant for interactive edits and coding-agent loops",
+        "family": "kimi-k2",
+        "id": "kimi-for-coding-highspeed",
+        "knowledge": "2025-01",
+        "last_updated": "2026-06-12",
+        "limit": {
+          "context": 262144,
+          "output": 32768
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "video"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi For Coding HighSpeed",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [],
+        "release_date": "2026-06-12",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
       "kimi-k2-thinking": {
         "attachment": false,
         "cost": {
@@ -88424,49 +89049,6 @@
         "temperature": true,
         "tool_call": true
       },
-      "deepseek-v3.1": {
-        "attachment": true,
-        "cost": {
-          "cache_read": 0.112,
-          "input": 0.56,
-          "output": 1.68
-        },
-        "description": "DeepSeek chat model for instruction following, coding, and analysis",
-        "family": "deepseek",
-        "id": "deepseek-v3.1",
-        "last_updated": "2025-08-21",
-        "limit": {
-          "context": 128000,
-          "output": 32768
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "DeepSeek V3.1",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "effort",
-            "values": [
-              "minimal",
-              "low",
-              "medium",
-              "high"
-            ]
-          }
-        ],
-        "release_date": "2025-08-21",
-        "structured_output": true,
-        "temperature": true,
-        "tool_call": true
-      },
       "deepseek-v3.2": {
         "attachment": true,
         "cost": {
@@ -88607,37 +89189,6 @@
         "temperature": true,
         "tool_call": true
       },
-      "devstral-small-2507": {
-        "attachment": false,
-        "cost": {
-          "input": 0.1,
-          "output": 0.3
-        },
-        "description": "Mistral coding agent model for repository tasks and software engineering workflows",
-        "family": "devstral",
-        "id": "devstral-small-2507",
-        "knowledge": "2025-05",
-        "last_updated": "2025-07-10",
-        "limit": {
-          "context": 131072,
-          "output": 128000
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Devstral Small",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-07-10",
-        "status": "deprecated",
-        "temperature": true,
-        "tool_call": true
-      },
       "fugu-ultra": {
         "attachment": true,
         "cost": {
@@ -91896,36 +92447,6 @@
         "temperature": true,
         "tool_call": false
       },
-      "llama-3-8b-instruct": {
-        "attachment": false,
-        "cost": {
-          "input": 0.04,
-          "output": 0.04
-        },
-        "description": "Open Llama instruction model for multilingual chat, reasoning, and coding",
-        "family": "llama",
-        "id": "llama-3-8b-instruct",
-        "last_updated": "2025-04-03",
-        "limit": {
-          "context": 8192,
-          "output": 8192
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Llama 3 8B Instruct",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-04-03",
-        "structured_output": true,
-        "temperature": true,
-        "tool_call": false
-      },
       "llama-3.1-70b-instruct": {
         "attachment": false,
         "cost": {
@@ -92475,16 +92996,16 @@
       "minimax-m3": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.12,
-          "input": 0.6,
-          "output": 2.4
+          "cache_read": 0.06,
+          "input": 0.3,
+          "output": 1.2
         },
         "description": "MiniMax multimodal model for long-context coding, perception, and agent planning",
         "family": "minimax",
         "id": "minimax-m3",
         "last_updated": "2026-06-01",
         "limit": {
-          "context": 512000,
+          "context": 524288,
           "output": 128000
         },
         "modalities": {
@@ -92959,37 +93480,6 @@
         "temperature": false,
         "tool_call": true
       },
-      "pixtral-large-latest": {
-        "attachment": true,
-        "cost": {
-          "input": 4,
-          "output": 12
-        },
-        "description": "Mistral's larger vision model for document-heavy image understanding and chat",
-        "family": "pixtral",
-        "id": "pixtral-large-latest",
-        "knowledge": "2024-11",
-        "last_updated": "2024-11-04",
-        "limit": {
-          "context": 128000,
-          "output": 128000
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Pixtral Large (latest)",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2024-11-01",
-        "temperature": true,
-        "tool_call": true
-      },
       "qwen-coder-plus": {
         "attachment": false,
         "cost": {
@@ -93431,37 +93921,6 @@
         "temperature": true,
         "tool_call": true
       },
-      "qwen3-4b-fp8": {
-        "attachment": false,
-        "cost": {
-          "input": 0.03,
-          "output": 0.03
-        },
-        "description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
-        "family": "qwen",
-        "id": "qwen3-4b-fp8",
-        "last_updated": "2025-04-28",
-        "limit": {
-          "context": 128000,
-          "output": 8192
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Qwen3 4B FP8",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-04-28",
-        "structured_output": true,
-        "temperature": true,
-        "tool_call": true
-      },
       "qwen3-coder-30b-a3b-instruct": {
         "attachment": false,
         "cost": {
@@ -93805,69 +94264,6 @@
         "temperature": true,
         "tool_call": true
       },
-      "qwen3-vl-30b-a3b-thinking": {
-        "attachment": true,
-        "cost": {
-          "input": 0.2,
-          "output": 1
-        },
-        "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
-        "family": "qwen",
-        "id": "qwen3-vl-30b-a3b-thinking",
-        "last_updated": "2025-10-02",
-        "limit": {
-          "context": 131072,
-          "output": 8192
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Qwen3 VL 30B A3B Thinking",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-10-02",
-        "structured_output": true,
-        "temperature": true,
-        "tool_call": true
-      },
-      "qwen3-vl-8b-instruct": {
-        "attachment": true,
-        "cost": {
-          "input": 0.08,
-          "output": 0.5
-        },
-        "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
-        "family": "qwen",
-        "id": "qwen3-vl-8b-instruct",
-        "last_updated": "2025-08-19",
-        "limit": {
-          "context": 131072,
-          "output": 8192
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Qwen3 VL 8B Instruct",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-08-19",
-        "structured_output": false,
-        "temperature": true,
-        "tool_call": false
-      },
       "qwen3-vl-flash": {
         "attachment": true,
         "cost": {
@@ -96974,16 +97370,16 @@
       "google/gemini-flash-latest": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.075,
-          "input": 0.3,
-          "input_audio": 1,
-          "output": 2.5
+          "cache_read": 0.15,
+          "input": 1.5,
+          "input_audio": 1.5,
+          "output": 9
         },
         "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
         "family": "gemini-flash",
         "id": "google/gemini-flash-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-19",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -96992,8 +97388,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -97014,7 +97410,7 @@
             ]
           }
         ],
-        "release_date": "2025-09-25",
+        "release_date": "2026-05-19",
         "structured_output": true,
         "temperature": true,
         "tool_call": true
@@ -97023,14 +97419,15 @@
         "attachment": true,
         "cost": {
           "cache_read": 0.025,
-          "input": 0.1,
-          "output": 0.4
+          "input": 0.25,
+          "input_audio": 0.5,
+          "output": 1.5
         },
         "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
         "family": "gemini-flash-lite",
         "id": "google/gemini-flash-lite-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-07",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -97039,8 +97436,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -97061,7 +97458,7 @@
             ]
           }
         ],
-        "release_date": "2025-09-25",
+        "release_date": "2026-05-07",
         "structured_output": true,
         "temperature": true,
         "tool_call": true
@@ -102925,6 +103322,48 @@
         "structured_output": true,
         "temperature": false,
         "tool_call": true
+      },
+      "kimi-k3": {
+        "attachment": true,
+        "description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
+        "family": "kimi-k3",
+        "id": "kimi-k3",
+        "interleaved": {
+          "field": "reasoning_content"
+        },
+        "last_updated": "2026-07-16",
+        "limit": {
+          "context": 1048576,
+          "output": 131072
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "video"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi K3",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-07-16",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
       }
     },
     "name": "Moonshot AI",
@@ -103260,6 +103699,48 @@
         "structured_output": true,
         "temperature": false,
         "tool_call": true
+      },
+      "kimi-k3": {
+        "attachment": true,
+        "description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
+        "family": "kimi-k3",
+        "id": "kimi-k3",
+        "interleaved": {
+          "field": "reasoning_content"
+        },
+        "last_updated": "2026-07-16",
+        "limit": {
+          "context": 1048576,
+          "output": 131072
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "video"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi K3",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-07-16",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
       }
     },
     "name": "Moonshot AI (China)",
@@ -113941,10 +114422,10 @@
         },
         "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
         "id": "google/gemini-flash-latest",
-        "last_updated": "2026-03-29",
+        "last_updated": "2026-05-19",
         "limit": {
-          "context": 1048756,
-          "input": 1048756,
+          "context": 1048576,
+          "input": 1048576,
           "output": 65536
         },
         "modalities": {
@@ -113970,9 +114451,9 @@
             ]
           }
         ],
-        "release_date": "2026-03-29",
-        "structured_output": false,
-        "tool_call": false
+        "release_date": "2026-05-19",
+        "structured_output": true,
+        "tool_call": true
       },
       "google/gemini-flash-lite-latest": {
         "attachment": true,
@@ -113983,7 +114464,7 @@
         },
         "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
         "id": "google/gemini-flash-lite-latest",
-        "last_updated": "2026-03-29",
+        "last_updated": "2026-05-07",
         "limit": {
           "context": 1048576,
           "input": 1048576,
@@ -114013,7 +114494,7 @@
             ]
           }
         ],
-        "release_date": "2026-03-29",
+        "release_date": "2026-05-07",
         "structured_output": true,
         "tool_call": true
       },
@@ -134713,94 +135194,6 @@
     ],
     "id": "ollama-cloud",
     "models": {
-      "cogito-2.1:671b": {
-        "attachment": false,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "cogito",
-        "id": "cogito-2.1:671b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 163840,
-          "output": 32000
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "cogito-2.1:671b",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "toggle"
-          }
-        ],
-        "release_date": "2025-11-19",
-        "status": "deprecated",
-        "tool_call": true
-      },
-      "deepseek-v3.1:671b": {
-        "attachment": false,
-        "description": "DeepSeek chat model for instruction following, coding, and analysis",
-        "family": "deepseek",
-        "id": "deepseek-v3.1:671b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 163840,
-          "output": 163840
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "deepseek-v3.1:671b",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "toggle"
-          }
-        ],
-        "release_date": "2025-08-21",
-        "tool_call": true
-      },
-      "deepseek-v3.2": {
-        "attachment": false,
-        "description": "DeepSeek chat model for instruction following, coding, and analysis",
-        "family": "deepseek",
-        "id": "deepseek-v3.2",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 163840,
-          "output": 65536
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "deepseek-v3.2",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "toggle"
-          }
-        ],
-        "release_date": "2025-06-15",
-        "tool_call": true
-      },
       "deepseek-v4-flash": {
         "attachment": false,
         "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops",
@@ -134873,161 +135266,6 @@
         "release_date": "2026-04-24",
         "tool_call": true
       },
-      "devstral-2:123b": {
-        "attachment": false,
-        "description": "Mistral coding agent model for repository tasks and software engineering workflows",
-        "family": "devstral",
-        "id": "devstral-2:123b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 262144
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "devstral-2:123b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-12-09",
-        "tool_call": true
-      },
-      "devstral-small-2:24b": {
-        "attachment": true,
-        "description": "Mistral coding agent model for repository tasks and software engineering workflows",
-        "family": "devstral",
-        "id": "devstral-small-2:24b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 262144
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "devstral-small-2:24b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-12-09",
-        "tool_call": true
-      },
-      "gemini-3-flash-preview": {
-        "attachment": true,
-        "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
-        "family": "gemini-flash",
-        "id": "gemini-3-flash-preview",
-        "knowledge": "2025-01",
-        "last_updated": "2026-04-08",
-        "limit": {
-          "context": 1048576,
-          "output": 65536
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "gemini-3-flash-preview",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "toggle"
-          }
-        ],
-        "release_date": "2025-12-17",
-        "tool_call": true
-      },
-      "gemma3:12b": {
-        "attachment": true,
-        "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
-        "family": "gemma",
-        "id": "gemma3:12b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 131072,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "gemma3:12b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2024-12-01",
-        "tool_call": false
-      },
-      "gemma3:27b": {
-        "attachment": true,
-        "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
-        "family": "gemma",
-        "id": "gemma3:27b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 131072,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "gemma3:27b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-07-27",
-        "tool_call": false
-      },
-      "gemma3:4b": {
-        "attachment": true,
-        "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
-        "family": "gemma",
-        "id": "gemma3:4b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 131072,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "gemma3:4b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2024-12-01",
-        "tool_call": false
-      },
       "gemma4:31b": {
         "attachment": true,
         "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
@@ -135059,97 +135297,6 @@
         "release_date": "2026-04-02",
         "tool_call": true
       },
-      "glm-4.6": {
-        "attachment": false,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "glm",
-        "id": "glm-4.6",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 202752,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "glm-4.6",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "toggle"
-          }
-        ],
-        "release_date": "2025-09-29",
-        "status": "deprecated",
-        "tool_call": true
-      },
-      "glm-4.7": {
-        "attachment": false,
-        "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
-        "family": "glm",
-        "id": "glm-4.7",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 202752,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "glm-4.7",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "toggle"
-          }
-        ],
-        "release_date": "2025-12-22",
-        "tool_call": true
-      },
-      "glm-5": {
-        "attachment": false,
-        "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
-        "family": "glm",
-        "id": "glm-5",
-        "interleaved": {
-          "field": "reasoning_content"
-        },
-        "last_updated": "2026-02-11",
-        "limit": {
-          "context": 202752,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "glm-5",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "toggle"
-          }
-        ],
-        "release_date": "2026-02-11",
-        "tool_call": true
-      },
       "glm-5.1": {
         "attachment": false,
         "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
@@ -135288,33 +135435,6 @@
         "release_date": "2025-08-05",
         "tool_call": true
       },
-      "kimi-k2-thinking": {
-        "attachment": false,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "kimi-thinking",
-        "id": "kimi-k2-thinking",
-        "knowledge": "2024-08",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 262144
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "kimi-k2-thinking",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-11-06",
-        "status": "deprecated",
-        "tool_call": true
-      },
       "kimi-k2.5": {
         "attachment": true,
         "description": "Kimi multimodal agent model for visual understanding, coding, and planning",
@@ -135408,83 +135528,6 @@
         "temperature": false,
         "tool_call": true
       },
-      "kimi-k2:1t": {
-        "attachment": false,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "kimi-k2",
-        "id": "kimi-k2:1t",
-        "knowledge": "2024-10",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 262144
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "kimi-k2:1t",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-07-11",
-        "status": "deprecated",
-        "tool_call": true
-      },
-      "minimax-m2": {
-        "attachment": false,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "minimax",
-        "id": "minimax-m2",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 204800,
-          "output": 128000
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "minimax-m2",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-10-23",
-        "status": "deprecated",
-        "tool_call": true
-      },
-      "minimax-m2.1": {
-        "attachment": false,
-        "description": "MiniMax model for chat, coding, office work, and agentic tasks",
-        "family": "minimax",
-        "id": "minimax-m2.1",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 204800,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "minimax-m2.1",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-12-23",
-        "tool_call": true
-      },
       "minimax-m2.5": {
         "attachment": false,
         "description": "MiniMax model for chat, coding, office work, and agentic tasks",
@@ -135582,81 +135625,6 @@
         "temperature": true,
         "tool_call": true
       },
-      "ministral-3:14b": {
-        "attachment": true,
-        "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads",
-        "family": "ministral",
-        "id": "ministral-3:14b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 128000
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "ministral-3:14b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2024-12-01",
-        "tool_call": true
-      },
-      "ministral-3:3b": {
-        "attachment": true,
-        "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads",
-        "family": "ministral",
-        "id": "ministral-3:3b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 128000
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "ministral-3:3b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2024-10-22",
-        "tool_call": true
-      },
-      "ministral-3:8b": {
-        "attachment": true,
-        "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads",
-        "family": "ministral",
-        "id": "ministral-3:8b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 128000
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "ministral-3:8b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2024-12-01",
-        "tool_call": true
-      },
       "mistral-large-3:675b": {
         "attachment": true,
         "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work",
@@ -135772,133 +135740,6 @@
         "temperature": true,
         "tool_call": true
       },
-      "qwen3-coder-next": {
-        "attachment": false,
-        "description": "Qwen coding model for software agents, repository edits, and code reasoning",
-        "family": "qwen",
-        "id": "qwen3-coder-next",
-        "last_updated": "2026-02-08",
-        "limit": {
-          "context": 262144,
-          "output": 65536
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "qwen3-coder-next",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2026-02-02",
-        "tool_call": true
-      },
-      "qwen3-coder:480b": {
-        "attachment": false,
-        "description": "Qwen coding model for software agents, repository edits, and code reasoning",
-        "family": "qwen",
-        "id": "qwen3-coder:480b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 65536
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "qwen3-coder:480b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-07-22",
-        "tool_call": true
-      },
-      "qwen3-next:80b": {
-        "attachment": false,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "qwen",
-        "id": "qwen3-next:80b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 32768
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "qwen3-next:80b",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-09-15",
-        "status": "deprecated",
-        "tool_call": true
-      },
-      "qwen3-vl:235b": {
-        "attachment": true,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "qwen",
-        "id": "qwen3-vl:235b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 32768
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "qwen3-vl:235b",
-        "open_weights": true,
-        "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-09-22",
-        "status": "deprecated",
-        "tool_call": true
-      },
-      "qwen3-vl:235b-instruct": {
-        "attachment": true,
-        "description": "Legacy model retained for compatibility with older integrations",
-        "family": "qwen",
-        "id": "qwen3-vl:235b-instruct",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 262144,
-          "output": 131072
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "qwen3-vl:235b-instruct",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-09-22",
-        "status": "deprecated",
-        "tool_call": true
-      },
       "qwen3.5:397b": {
         "attachment": true,
         "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
@@ -135931,30 +135772,6 @@
         ],
         "release_date": "2026-02-15",
         "tool_call": true
-      },
-      "rnj-1:8b": {
-        "attachment": false,
-        "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads",
-        "family": "rnj",
-        "id": "rnj-1:8b",
-        "last_updated": "2026-01-19",
-        "limit": {
-          "context": 32768,
-          "output": 4096
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "rnj-1:8b",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-12-06",
-        "tool_call": true
       }
     },
     "name": "Ollama Cloud",
@@ -143590,9 +143407,6 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -143603,8 +143417,6 @@
             ]
           },
           {
-            "max": 127999,
-            "min": 1024,
             "type": "budget_tokens"
           }
         ],
@@ -143662,9 +143474,6 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -143692,7 +143501,8 @@
         "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
         "family": "claude-opus",
         "id": "anthropic/claude-opus-4.7-fast",
-        "last_updated": "2026-05-12",
+        "knowledge": "2026-01-31",
+        "last_updated": "2026-04-16",
         "limit": {
           "context": 1000000,
           "output": 128000
@@ -143711,9 +143521,6 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -143725,7 +143532,7 @@
             ]
           }
         ],
-        "release_date": "2026-05-12",
+        "release_date": "2026-04-16",
         "structured_output": true,
         "temperature": false,
         "tool_call": true
@@ -143761,9 +143568,6 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -143791,7 +143595,8 @@
         "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
         "family": "claude-opus",
         "id": "anthropic/claude-opus-4.8-fast",
-        "last_updated": "2026-05-27",
+        "knowledge": "2026-01",
+        "last_updated": "2026-05-28",
         "limit": {
           "context": 1000000,
           "output": 128000
@@ -143810,9 +143615,6 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -143824,7 +143626,7 @@
             ]
           }
         ],
-        "release_date": "2026-05-27",
+        "release_date": "2026-05-28",
         "structured_output": true,
         "temperature": false,
         "tool_call": true
@@ -144004,9 +143806,6 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -144015,11 +143814,6 @@
               "high",
               "max"
             ]
-          },
-          {
-            "max": 127999,
-            "min": 1024,
-            "type": "budget_tokens"
           }
         ],
         "release_date": "2026-02-17",
@@ -144074,36 +143868,6 @@
         "temperature": false,
         "tool_call": true
       },
-      "arcee-ai/coder-large": {
-        "attachment": false,
-        "cost": {
-          "input": 0.5,
-          "output": 0.8
-        },
-        "description": "Coding model for repository understanding, refactors, and agentic engineering tasks",
-        "id": "arcee-ai/coder-large",
-        "knowledge": "2025-03-31",
-        "last_updated": "2025-05-05",
-        "limit": {
-          "context": 32768,
-          "output": 32768
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Coder Large",
-        "open_weights": false,
-        "reasoning": false,
-        "release_date": "2025-05-05",
-        "structured_output": false,
-        "temperature": true,
-        "tool_call": false
-      },
       "arcee-ai/trinity-large-thinking": {
         "attachment": false,
         "cost": {
@@ -144666,8 +144430,8 @@
         "attachment": false,
         "cost": {
           "cache_read": 0.135,
-          "input": 0.24,
-          "output": 0.9
+          "input": 0.27,
+          "output": 1.12
         },
         "description": "DeepSeek chat model for instruction following, coding, and analysis",
         "family": "deepseek",
@@ -144676,7 +144440,7 @@
         "last_updated": "2025-03-24",
         "limit": {
           "context": 163840,
-          "output": 16384
+          "output": 65536
         },
         "modalities": {
           "input": [
@@ -144831,9 +144595,9 @@
       "deepseek/deepseek-v3.1-terminus": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.13,
+          "cache_read": 0.135,
           "input": 0.27,
-          "output": 0.95
+          "output": 1
         },
         "description": "DeepSeek chat model for instruction following, coding, and analysis",
         "family": "deepseek",
@@ -144841,7 +144605,7 @@
         "knowledge": "2025-03-31",
         "last_updated": "2025-09-22",
         "limit": {
-          "context": 163840,
+          "context": 131072,
           "output": 32768
         },
         "modalities": {
@@ -144868,9 +144632,9 @@
       "deepseek/deepseek-v3.2": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.02145,
-          "input": 0.2145,
-          "output": 0.32175
+          "cache_read": 0.1345,
+          "input": 0.269,
+          "output": 0.4
         },
         "description": "DeepSeek chat model for instruction following, coding, and analysis",
         "family": "deepseek",
@@ -144878,8 +144642,8 @@
         "knowledge": "2024-07",
         "last_updated": "2025-12-01",
         "limit": {
-          "context": 131072,
-          "output": 64000
+          "context": 163840,
+          "output": 65536
         },
         "modalities": {
           "input": [
@@ -144941,9 +144705,9 @@
       "deepseek/deepseek-v4-flash": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.018,
-          "input": 0.09,
-          "output": 0.18
+          "cache_read": 0.02,
+          "input": 0.098,
+          "output": 0.196
         },
         "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work",
         "family": "deepseek-flash",
@@ -144969,9 +144733,6 @@
         "open_weights": true,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -145016,9 +144777,6 @@
         "open_weights": true,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -145095,7 +144853,7 @@
         "last_updated": "2025-08-26",
         "limit": {
           "context": 32768,
-          "output": 32768
+          "output": 8192
         },
         "modalities": {
           "input": [
@@ -145569,7 +145327,7 @@
         "last_updated": "2026-06-30",
         "limit": {
           "context": 65536,
-          "output": 66000
+          "output": 65536
         },
         "modalities": {
           "input": [
@@ -145896,8 +145654,9 @@
       "google/gemma-3-27b-it": {
         "attachment": true,
         "cost": {
+          "cache_read": 0.04,
           "input": 0.08,
-          "output": 0.16
+          "output": 0.45
         },
         "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
         "family": "gemma",
@@ -145906,7 +145665,7 @@
         "last_updated": "2025-03-12",
         "limit": {
           "context": 131072,
-          "output": 16384
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -145991,8 +145750,8 @@
       "google/gemma-4-26b-a4b-it": {
         "attachment": true,
         "cost": {
-          "input": 0.06,
-          "output": 0.33
+          "input": 0.1,
+          "output": 0.3
         },
         "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
         "family": "gemma",
@@ -146000,7 +145759,7 @@
         "last_updated": "2026-04-02",
         "limit": {
           "context": 262144,
-          "output": 262144
+          "output": 256000
         },
         "modalities": {
           "input": [
@@ -146065,8 +145824,9 @@
       "google/gemma-4-31b-it": {
         "attachment": true,
         "cost": {
-          "input": 0.06,
-          "output": 0.35
+          "cache_read": 0.12,
+          "input": 0.22,
+          "output": 0.55
         },
         "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
         "family": "gemma",
@@ -146074,7 +145834,7 @@
         "last_updated": "2026-04-02",
         "limit": {
           "context": 262144,
-          "output": 8192
+          "output": 262144
         },
         "modalities": {
           "input": [
@@ -146653,8 +146413,9 @@
       "meta-llama/llama-3.1-8b-instruct": {
         "attachment": false,
         "cost": {
-          "input": 0.02,
-          "output": 0.03
+          "cache_read": 0.025,
+          "input": 0.05,
+          "output": 0.08
         },
         "description": "Open Llama instruction model for multilingual chat, reasoning, and coding",
         "family": "llama",
@@ -146663,7 +146424,7 @@
         "last_updated": "2024-07-23",
         "limit": {
           "context": 131072,
-          "output": 16384
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -146809,8 +146570,8 @@
       "meta-llama/llama-3.3-70b-instruct": {
         "attachment": false,
         "cost": {
-          "input": 0.1,
-          "output": 0.32
+          "input": 0.13,
+          "output": 0.4
         },
         "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting",
         "family": "llama",
@@ -146819,7 +146580,7 @@
         "last_updated": "2024-12-06",
         "limit": {
           "context": 131072,
-          "output": 16384
+          "output": 128000
         },
         "modalities": {
           "input": [
@@ -146964,6 +146725,53 @@
         "temperature": true,
         "tool_call": false
       },
+      "meta/muse-spark-1.1": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.15,
+          "input": 1.25,
+          "output": 4.25
+        },
+        "description": "Open Llama multimodal model for image understanding and text reasoning",
+        "family": "muse",
+        "id": "meta/muse-spark-1.1",
+        "last_updated": "2026-07-09",
+        "limit": {
+          "context": 1048576,
+          "output": 1048576
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "video",
+            "pdf",
+            "audio"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Muse Spark 1.1",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "minimal",
+              "low",
+              "medium",
+              "high",
+              "xhigh"
+            ]
+          }
+        ],
+        "release_date": "2026-04-08",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
+      },
       "microsoft/phi-4": {
         "attachment": false,
         "cost": {
@@ -147060,7 +146868,7 @@
       "minimax/minimax-m1": {
         "attachment": false,
         "cost": {
-          "input": 0.4,
+          "input": 0.55,
           "output": 2.2
         },
         "description": "MiniMax model for chat, coding, office work, and agentic tasks",
@@ -147227,8 +147035,9 @@
       "minimax/minimax-m2.7": {
         "attachment": false,
         "cost": {
-          "input": 0.24,
-          "output": 0.96
+          "cache_read": 0.06,
+          "input": 0.3,
+          "output": 1.2
         },
         "description": "Open MiniMax flagship for coding agents, office automation, and complex environments",
         "family": "minimax",
@@ -147236,7 +147045,7 @@
         "last_updated": "2026-03-18",
         "limit": {
           "context": 204800,
-          "output": 196608
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -147268,7 +147077,7 @@
         "last_updated": "2026-06-01",
         "limit": {
           "context": 1048576,
-          "output": 131072
+          "output": 512000
         },
         "modalities": {
           "input": [
@@ -147665,7 +147474,7 @@
         "attachment": false,
         "cost": {
           "input": 0.02,
-          "output": 0.03
+          "output": 0.04
         },
         "description": "Efficient Mistral-NVIDIA open model for multilingual chat and local deployment",
         "family": "mistral-nemo",
@@ -147674,7 +147483,7 @@
         "last_updated": "2024-07-01",
         "limit": {
           "context": 131072,
-          "output": 131072
+          "output": 16384
         },
         "modalities": {
           "input": [
@@ -147833,8 +147642,9 @@
       "mistralai/mistral-small-3.2-24b-instruct": {
         "attachment": true,
         "cost": {
-          "input": 0.075,
-          "output": 0.2
+          "cache_read": 0.01,
+          "input": 0.1,
+          "output": 0.3
         },
         "description": "Efficient Mistral model for fast chat, extraction, and production assistants",
         "family": "mistral-small",
@@ -147842,7 +147652,7 @@
         "knowledge": "2023-10-31",
         "last_updated": "2025-06-20",
         "limit": {
-          "context": 128000,
+          "context": 131072,
           "output": 16384
         },
         "modalities": {
@@ -147993,7 +147803,6 @@
       "moonshotai/kimi-k2-thinking": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.15,
           "input": 0.6,
           "output": 2.5
         },
@@ -148007,7 +147816,7 @@
         "last_updated": "2025-11-06",
         "limit": {
           "context": 262144,
-          "output": 100352
+          "output": 262144
         },
         "modalities": {
           "input": [
@@ -148029,9 +147838,9 @@
       "moonshotai/kimi-k2.5": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.203,
-          "input": 0.375,
-          "output": 2.025
+          "cache_read": 0.095,
+          "input": 0.57,
+          "output": 2.85
         },
         "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work",
         "family": "kimi-k2",
@@ -148043,7 +147852,7 @@
         "last_updated": "2026-01",
         "limit": {
           "context": 262144,
-          "output": 256000
+          "output": 262144
         },
         "modalities": {
           "input": [
@@ -148066,9 +147875,9 @@
       "moonshotai/kimi-k2.6": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.15,
-          "input": 0.66,
-          "output": 3.41
+          "cache_read": 0.16,
+          "input": 0.95,
+          "output": 4
         },
         "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context",
         "family": "kimi-k2",
@@ -148103,9 +147912,9 @@
       "moonshotai/kimi-k2.7-code": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.149,
-          "input": 0.719,
-          "output": 3.49
+          "cache_read": 0.16,
+          "input": 0.75,
+          "output": 3.5
         },
         "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
         "family": "kimi-k2",
@@ -148134,6 +147943,46 @@
         "temperature": true,
         "tool_call": true
       },
+      "moonshotai/kimi-k3": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.3,
+          "input": 3,
+          "output": 15
+        },
+        "description": "Kimi multimodal agent model for visual understanding, coding, and planning",
+        "family": "kimi-k3",
+        "id": "moonshotai/kimi-k3",
+        "last_updated": "2026-07-16",
+        "limit": {
+          "context": 1048576,
+          "output": 1048576
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi K3",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "max"
+            ]
+          }
+        ],
+        "release_date": "2026-07-16",
+        "structured_output": true,
+        "temperature": false,
+        "tool_call": true
+      },
       "morph/morph-v3-fast": {
         "attachment": false,
         "cost": {
@@ -148559,8 +148408,9 @@
       "nvidia/nemotron-3-super-120b-a12b": {
         "attachment": false,
         "cost": {
-          "input": 0.08,
-          "output": 0.45
+          "cache_read": 0.06,
+          "input": 0.21,
+          "output": 0.455
         },
         "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads",
         "family": "nemotron",
@@ -148643,9 +148493,9 @@
       "nvidia/nemotron-3-ultra-550b-a55b": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.1,
-          "input": 0.5,
-          "output": 2.2
+          "cache_read": 0.2,
+          "input": 0.6,
+          "output": 3.6
         },
         "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy",
         "family": "nemotron",
@@ -149144,6 +148994,7 @@
       "openai/gpt-4o": {
         "attachment": true,
         "cost": {
+          "cache_read": 1.25,
           "input": 2.5,
           "output": 10
         },
@@ -149783,7 +149634,7 @@
       "openai/gpt-5.1-chat": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.13,
+          "cache_read": 0.125,
           "input": 1.25,
           "output": 10
         },
@@ -149794,7 +149645,7 @@
         "last_updated": "2025-11-13",
         "limit": {
           "context": 128000,
-          "output": 32000
+          "output": 16384
         },
         "modalities": {
           "input": [
@@ -149817,7 +149668,7 @@
       "openai/gpt-5.1-codex": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.13,
+          "cache_read": 0.125,
           "input": 1.25,
           "output": 10
         },
@@ -150184,6 +150035,7 @@
           {
             "type": "effort",
             "values": [
+              "none",
               "low",
               "medium",
               "high",
@@ -150912,8 +150764,8 @@
       "openai/gpt-oss-120b": {
         "attachment": false,
         "cost": {
-          "input": 0.03,
-          "output": 0.15
+          "input": 0.037,
+          "output": 0.17
         },
         "description": "Open GPT reasoning model for self-hosted agents and controllable deployments",
         "family": "gpt-oss",
@@ -150952,8 +150804,9 @@
       "openai/gpt-oss-20b": {
         "attachment": false,
         "cost": {
-          "input": 0.029,
-          "output": 0.14
+          "cache_read": 0.03,
+          "input": 0.03,
+          "output": 0.13
         },
         "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
         "family": "gpt-oss",
@@ -152132,8 +151985,9 @@
       "qwen/qwen2.5-vl-72b-instruct": {
         "attachment": true,
         "cost": {
-          "input": 0.25,
-          "output": 0.75
+          "cache_read": 0.4,
+          "input": 0.8,
+          "output": 1
         },
         "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
         "family": "qwen",
@@ -152340,8 +152194,8 @@
       "qwen/qwen3-30b-a3b-instruct-2507": {
         "attachment": false,
         "cost": {
-          "input": 0.04815,
-          "output": 0.19305
+          "input": 0.1,
+          "output": 0.3
         },
         "description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
         "family": "qwen",
@@ -152349,7 +152203,7 @@
         "knowledge": "2025-06-30",
         "last_updated": "2025-07-29",
         "limit": {
-          "context": 131072,
+          "context": 262144,
           "output": 32000
         },
         "modalities": {
@@ -152475,8 +152329,9 @@
       "qwen/qwen3-coder": {
         "attachment": false,
         "cost": {
-          "input": 0.22,
-          "output": 1.8
+          "cache_read": 0.1,
+          "input": 0.3,
+          "output": 1
         },
         "description": "Qwen coding model for software agents, repository edits, and code reasoning",
         "family": "qwen",
@@ -152735,7 +152590,8 @@
       "qwen/qwen3-next-80b-a3b-instruct": {
         "attachment": false,
         "cost": {
-          "input": 0.09,
+          "cache_read": 0.07,
+          "input": 0.1,
           "output": 1.1
         },
         "description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
@@ -152745,7 +152601,7 @@
         "last_updated": "2025-09",
         "limit": {
           "context": 262144,
-          "output": 16384
+          "output": 262144
         },
         "modalities": {
           "input": [
@@ -152829,9 +152685,9 @@
       "qwen/qwen3-vl-235b-a22b-instruct": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.11,
-          "input": 0.2,
-          "output": 0.88
+          "cache_read": 0.1,
+          "input": 0.21,
+          "output": 1.9
         },
         "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
         "family": "qwen",
@@ -152839,8 +152695,8 @@
         "knowledge": "2025-03-31",
         "last_updated": "2025-09-23",
         "limit": {
-          "context": 262144,
-          "output": 16384
+          "context": 131072,
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -153063,7 +152919,7 @@
         "last_updated": "2026-02-23",
         "limit": {
           "context": 262144,
-          "output": 262144
+          "output": 65536
         },
         "modalities": {
           "input": [
@@ -153128,7 +152984,6 @@
       "qwen/qwen3.5-35b-a3b": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.05,
           "input": 0.14,
           "output": 1
         },
@@ -153138,7 +152993,7 @@
         "last_updated": "2026-02-23",
         "limit": {
           "context": 262144,
-          "output": 81920
+          "output": 262144
         },
         "modalities": {
           "input": [
@@ -153166,17 +153021,17 @@
       "qwen/qwen3.5-397b-a17b": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.111,
-          "input": 0.385,
-          "output": 2.45
+          "cache_read": 0.225,
+          "input": 0.45,
+          "output": 3
         },
         "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks",
         "family": "qwen",
         "id": "qwen/qwen3.5-397b-a17b",
         "last_updated": "2026-02-15",
         "limit": {
-          "context": 256000,
-          "output": 64000
+          "context": 262144,
+          "output": 65536
         },
         "modalities": {
           "input": [
@@ -153369,8 +153224,8 @@
       "qwen/qwen3.6-27b": {
         "attachment": true,
         "cost": {
-          "input": 0.289,
-          "output": 2.4
+          "input": 0.45,
+          "output": 2.7
         },
         "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
         "family": "qwen",
@@ -153378,7 +153233,7 @@
         "last_updated": "2026-04-22",
         "limit": {
           "context": 262144,
-          "output": 131072
+          "output": 65536
         },
         "modalities": {
           "input": [
@@ -153572,10 +153427,10 @@
       "qwen/qwen3.7-max": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.25,
-          "cache_write": 1.5625,
-          "input": 1.25,
-          "output": 3.75
+          "cache_read": 0.295,
+          "cache_write": 1.84375,
+          "input": 1.475,
+          "output": 4.425
         },
         "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks",
         "family": "qwen",
@@ -154022,9 +153877,9 @@
       "tencent/hy3": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.035,
-          "input": 0.14,
-          "output": 0.58
+          "cache_read": 0.05,
+          "input": 0.2,
+          "output": 0.8
         },
         "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
         "family": "hy3",
@@ -154424,7 +154279,9 @@
             "type": "effort",
             "values": [
               "low",
-              "high"
+              "medium",
+              "high",
+              "xhigh"
             ]
           }
         ],
@@ -154589,8 +154446,8 @@
       "xiaomi/mimo-v2.5": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.028,
-          "input": 0.105,
+          "cache_read": 0.0028,
+          "input": 0.14,
           "output": 0.28
         },
         "description": "Open MiMo model for multimodal coding agents and long-context automation",
@@ -154784,9 +154641,9 @@
       "z-ai/glm-4.6": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.08,
-          "input": 0.43,
-          "output": 1.75
+          "cache_read": 0.1,
+          "input": 0.5,
+          "output": 2
         },
         "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks",
         "family": "glm",
@@ -154794,8 +154651,8 @@
         "knowledge": "2025-04",
         "last_updated": "2025-09-30",
         "limit": {
-          "context": 200000,
-          "output": 16384
+          "context": 202752,
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -154888,8 +154745,7 @@
       "z-ai/glm-4.7-flash": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.01,
-          "input": 0.06,
+          "input": 0.0605,
           "output": 0.4
         },
         "description": "Budget GLM lane for fast coding help, routing, and everyday automation",
@@ -154901,8 +154757,8 @@
         "knowledge": "2025-04",
         "last_updated": "2026-01-19",
         "limit": {
-          "context": 202752,
-          "output": 16384
+          "context": 200000,
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -154924,9 +154780,9 @@
       "z-ai/glm-5": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.12,
-          "input": 0.6,
-          "output": 1.92
+          "cache_read": 0.19,
+          "input": 0.95,
+          "output": 3.15
         },
         "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows",
         "family": "glm",
@@ -154937,7 +154793,7 @@
         "last_updated": "2026-02-12",
         "limit": {
           "context": 202752,
-          "output": 128000
+          "output": 202752
         },
         "modalities": {
           "input": [
@@ -154971,7 +154827,7 @@
         },
         "last_updated": "2026-03-16",
         "limit": {
-          "context": 262144,
+          "context": 202752,
           "output": 131072
         },
         "modalities": {
@@ -155029,9 +154885,9 @@
       "z-ai/glm-5.2": {
         "attachment": false,
         "cost": {
-          "cache_read": 0.169,
-          "input": 0.91,
-          "output": 2.86
+          "cache_read": 0.17342,
+          "input": 0.9338,
+          "output": 2.9348
         },
         "description": "Open flagship GLM for long-horizon coding agents and million-token context work",
         "family": "glm",
@@ -155042,7 +154898,7 @@
         "last_updated": "2026-06-13",
         "limit": {
           "context": 1048576,
-          "output": 128000
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -155223,9 +155079,6 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
@@ -155253,6 +155106,7 @@
         "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control",
         "family": "claude-sonnet",
         "id": "~anthropic/claude-sonnet-latest",
+        "knowledge": "2026-01-31",
         "last_updated": "2026-04-27",
         "limit": {
           "context": 1000000,
@@ -155272,22 +155126,15 @@
         "open_weights": false,
         "reasoning": true,
         "reasoning_options": [
-          {
-            "type": "toggle"
-          },
           {
             "type": "effort",
             "values": [
               "low",
               "medium",
               "high",
+              "xhigh",
               "max"
             ]
-          },
-          {
-            "max": 127999,
-            "min": 1024,
-            "type": "budget_tokens"
           }
         ],
         "release_date": "2026-04-27",
@@ -155349,13 +155196,30 @@
         "cost": {
           "cache_read": 0.2,
           "cache_write": 0.375,
+          "context_over_200k": {
+            "cache_read": 0.4,
+            "input": 4,
+            "output": 18
+          },
           "input": 2,
           "output": 12,
-          "reasoning": 12
+          "reasoning": 12,
+          "tiers": [
+            {
+              "cache_read": 0.4,
+              "input": 4,
+              "output": 18,
+              "tier": {
+                "size": 200000,
+                "type": "context"
+              }
+            }
+          ]
         },
         "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis",
         "family": "gemini-pro",
         "id": "~google/gemini-pro-latest",
+        "knowledge": "2025-01",
         "last_updated": "2026-04-27",
         "limit": {
           "context": 1048576,
@@ -155394,16 +155258,16 @@
       "~moonshotai/kimi-latest": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.15,
-          "input": 0.66,
-          "output": 3.41
+          "cache_read": 0.3,
+          "input": 3,
+          "output": 15
         },
         "description": "Kimi multimodal agent model for visual understanding, coding, and planning",
         "family": "kimi",
         "id": "~moonshotai/kimi-latest",
         "last_updated": "2026-04-27",
         "limit": {
-          "context": 262144,
+          "context": 1048576,
           "output": 262144
         },
         "modalities": {
@@ -155418,10 +155282,17 @@
         "name": "MoonshotAI Kimi Latest",
         "open_weights": false,
         "reasoning": true,
-        "reasoning_options": [],
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "max"
+            ]
+          }
+        ],
         "release_date": "2026-04-27",
         "structured_output": true,
-        "temperature": true,
+        "temperature": false,
         "tool_call": true
       },
       "~openai/gpt-latest": {
@@ -155462,7 +155333,8 @@
               "low",
               "medium",
               "high",
-              "xhigh"
+              "xhigh",
+              "max"
             ]
           }
         ],
@@ -155549,7 +155421,6 @@
           {
             "type": "effort",
             "values": [
-              "none",
               "low",
               "medium",
               "high"
@@ -156532,16 +156403,16 @@
       "google/gemini-flash-latest": {
         "attachment": true,
         "cost": {
-          "cache_read": 0.075,
-          "input": 0.5,
-          "input_audio": 1,
-          "output": 3
+          "cache_read": 0.15,
+          "input": 1.5,
+          "input_audio": 1.5,
+          "output": 9
         },
         "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
         "family": "gemini-flash",
         "id": "google/gemini-flash-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-19",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -156550,8 +156421,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -156561,8 +156432,18 @@
         "name": "Gemini Flash Latest",
         "open_weights": false,
         "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-09-25",
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "minimal",
+              "low",
+              "medium",
+              "high"
+            ]
+          }
+        ],
+        "release_date": "2026-05-19",
         "structured_output": true,
         "temperature": true,
         "tool_call": true
@@ -156572,13 +156453,14 @@
         "cost": {
           "cache_read": 0.025,
           "input": 0.25,
+          "input_audio": 0.5,
           "output": 1.5
         },
         "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
         "family": "gemini-flash-lite",
         "id": "google/gemini-flash-lite-latest",
         "knowledge": "2025-01",
-        "last_updated": "2025-09-25",
+        "last_updated": "2026-05-07",
         "limit": {
           "context": 1048576,
           "output": 65536
@@ -156587,8 +156469,8 @@
           "input": [
             "text",
             "image",
-            "audio",
             "video",
+            "audio",
             "pdf"
           ],
           "output": [
@@ -156598,8 +156480,18 @@
         "name": "Gemini Flash-Lite Latest",
         "open_weights": false,
         "reasoning": true,
-        "reasoning_options": [],
-        "release_date": "2025-09-25",
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "minimal",
+              "low",
+              "medium",
+              "high"
+            ]
+          }
+        ],
+        "release_date": "2026-05-07",
         "structured_output": true,
         "temperature": true,
         "tool_call": true
@@ -160256,7 +160148,7 @@
         "last_updated": "2025-06-30",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -160300,7 +160192,7 @@
         "last_updated": "2026-02-25",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -160433,7 +160325,7 @@
         "last_updated": "2025-03-31",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -160522,7 +160414,7 @@
         "last_updated": "2025-03-31",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -160565,7 +160457,7 @@
         "last_updated": "2025-07-31",
         "limit": {
           "context": 262144,
-          "output": 4096
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -160611,7 +160503,7 @@
         "last_updated": "2025-04-28",
         "limit": {
           "context": 131072,
-          "output": 4096
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -160656,7 +160548,7 @@
         "last_updated": "2026-02-23",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -160702,7 +160594,7 @@
         "last_updated": "2026-04-22",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -161691,7 +161583,7 @@
         "last_updated": "2026-05-31",
         "limit": {
           "context": 262144,
-          "output": 4096
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -161735,7 +161627,7 @@
         "last_updated": "2025-02-28",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -161778,7 +161670,7 @@
         "last_updated": "2026-05-31",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -161822,7 +161714,7 @@
         "last_updated": "2026-04-02",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -161868,7 +161760,7 @@
         "last_updated": "2026-04-02",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -161915,7 +161807,7 @@
         "last_updated": "2026-04-02",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -162728,7 +162620,7 @@
         "last_updated": "2024-12-06",
         "limit": {
           "context": 131072,
-          "output": 4096
+          "output": 131072
         },
         "modalities": {
           "input": [
@@ -162817,7 +162709,7 @@
         "last_updated": "2023-04-30",
         "limit": {
           "context": 32768,
-          "output": 4096
+          "output": 32768
         },
         "modalities": {
           "input": [
@@ -168278,38 +168170,6 @@
     ],
     "id": "privatemode-ai",
     "models": {
-      "gemma-3-27b": {
-        "attachment": true,
-        "cost": {
-          "input": 0,
-          "output": 0
-        },
-        "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
-        "family": "gemma",
-        "id": "gemma-3-27b",
-        "knowledge": "2024-08",
-        "last_updated": "2025-03-12",
-        "limit": {
-          "context": 128000,
-          "output": 8192
-        },
-        "modalities": {
-          "input": [
-            "text",
-            "image"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Gemma 3 27B",
-        "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-03-12",
-        "structured_output": true,
-        "temperature": true,
-        "tool_call": true
-      },
       "gpt-oss-120b": {
         "attachment": false,
         "cost": {
@@ -168351,33 +168211,39 @@
         "temperature": true,
         "tool_call": true
       },
-      "qwen3-coder-30b-a3b": {
-        "attachment": false,
+      "kimi-k2.6": {
+        "attachment": true,
         "cost": {
           "input": 0,
           "output": 0
         },
-        "description": "Qwen coding model for software agents, repository edits, and code reasoning",
-        "family": "qwen",
-        "id": "qwen3-coder-30b-a3b",
-        "knowledge": "2025-04",
-        "last_updated": "2025-04",
+        "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context",
+        "family": "kimi-k2",
+        "id": "kimi-k2.6",
+        "knowledge": "2025-01",
+        "last_updated": "2026-04-21",
         "limit": {
-          "context": 128000,
-          "output": 32768
+          "context": 262144,
+          "output": 262144
         },
         "modalities": {
           "input": [
-            "text"
+            "text",
+            "image"
           ],
           "output": [
             "text"
           ]
         },
-        "name": "Qwen3-Coder 30B-A3B",
+        "name": "Kimi K2.6",
         "open_weights": true,
-        "reasoning": false,
-        "release_date": "2025-04",
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          }
+        ],
+        "release_date": "2026-04-21",
         "structured_output": true,
         "temperature": true,
         "tool_call": true
@@ -168413,6 +168279,36 @@
         "temperature": true,
         "tool_call": false
       },
+      "voxtral-mini-3b": {
+        "attachment": true,
+        "cost": {
+          "input": 0,
+          "output": 0
+        },
+        "description": "Speech-to-text model for audio transcription, translation, and audio understanding",
+        "family": "voxtral",
+        "id": "voxtral-mini-3b",
+        "last_updated": "2025-07",
+        "limit": {
+          "context": 32000,
+          "output": 32000
+        },
+        "modalities": {
+          "input": [
+            "audio"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Voxtral Mini 3B",
+        "open_weights": true,
+        "reasoning": false,
+        "release_date": "2025-07",
+        "structured_output": false,
+        "temperature": true,
+        "tool_call": false
+      },
       "whisper-large-v3": {
         "attachment": true,
         "cost": {
@@ -183395,6 +183291,67 @@
     "name": "The Grid AI",
     "npm": "@ai-sdk/openai-compatible"
   },
+  "thinkingmachines": {
+    "api": "https://tinker.thinkingmachines.dev/services/tinker-prod/oai/api/v1",
+    "doc": "https://tinker-docs.thinkingmachines.ai/tinker/compatible-apis/openai/",
+    "env": [
+      "TINKER_API_KEY"
+    ],
+    "id": "thinkingmachines",
+    "models": {
+      "inkling": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.748,
+          "input": 3.74,
+          "output": 9.36
+        },
+        "description": "Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio",
+        "family": "ling",
+        "id": "inkling",
+        "interleaved": {
+          "field": "reasoning_content"
+        },
+        "last_updated": "2026-07-15",
+        "limit": {
+          "context": 256000,
+          "output": 256000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "audio",
+            "pdf"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Inkling",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "none",
+              "minimal",
+              "low",
+              "medium",
+              "high",
+              "xhigh"
+            ]
+          }
+        ],
+        "release_date": "2026-07-15",
+        "temperature": true,
+        "tool_call": true
+      }
+    },
+    "name": "Thinking Machines",
+    "npm": "@ai-sdk/openai-compatible"
+  },
   "tinfoil": {
     "api": "https://inference.tinfoil.sh/v1",
     "doc": "https://docs.tinfoil.sh",
@@ -186556,44 +186513,6 @@
     ],
     "id": "venice",
     "models": {
-      "aion-labs-aion-2-0": {
-        "attachment": false,
-        "cost": {
-          "cache_read": 0.25,
-          "input": 1,
-          "output": 2
-        },
-        "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use",
-        "id": "aion-labs-aion-2-0",
-        "last_updated": "2026-06-11",
-        "limit": {
-          "context": 128000,
-          "output": 32768
-        },
-        "modalities": {
-          "input": [
-            "text"
-          ],
-          "output": [
-            "text"
-          ]
-        },
-        "name": "Aion 2.0",
-        "open_weights": false,
-        "reasoning": true,
-        "reasoning_options": [
-          {
-            "type": "effort",
-            "values": [
-              "low",
-              "medium",
-              "high"
-            ]
-          }
-        ],
-        "release_date": "2026-03-24",
-        "tool_call": false
-      },
       "aion-labs-aion-3-0": {
         "attachment": false,
         "cost": {
@@ -186701,7 +186620,18 @@
         "name": "Claude Fable 5",
         "open_weights": false,
         "reasoning": true,
-        "reasoning_options": [],
+        "reasoning_options": [
+          {
+            "type": "effort",
+            "values": [
+              "low",
+              "medium",
+              "high",
+              "xhigh",
+              "max"
+            ]
+          }
+        ],
         "release_date": "2026-06-10",
         "structured_output": true,
         "temperature": false,
@@ -191656,6 +191586,54 @@
         "temperature": true,
         "tool_call": true
       },
+      "anthropic/claude-opus-4.7-fast": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 3,
+          "cache_write": 37.5,
+          "input": 30,
+          "output": 150
+        },
+        "description": "Stronger Opus tier for advanced software work and high-stakes reasoning",
+        "family": "claude-opus",
+        "id": "anthropic/claude-opus-4.7-fast",
+        "knowledge": "2026-01-31",
+        "last_updated": "2026-04-16",
+        "limit": {
+          "context": 1000000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "pdf"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Claude Opus 4.7 (Fast)",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "low",
+              "medium",
+              "high",
+              "xhigh"
+            ]
+          }
+        ],
+        "release_date": "2026-04-16",
+        "temperature": true,
+        "tool_call": true
+      },
       "anthropic/claude-opus-4.8": {
         "attachment": true,
         "cost": {
@@ -191704,6 +191682,54 @@
         "temperature": true,
         "tool_call": true
       },
+      "anthropic/claude-opus-4.8-fast": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 1,
+          "cache_write": 12.5,
+          "input": 10,
+          "output": 50
+        },
+        "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents",
+        "family": "claude-opus",
+        "id": "anthropic/claude-opus-4.8-fast",
+        "knowledge": "2026-01",
+        "last_updated": "2026-05-28",
+        "limit": {
+          "context": 1000000,
+          "output": 128000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "pdf"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Claude Opus 4.8 (Fast)",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [
+          {
+            "type": "toggle"
+          },
+          {
+            "type": "effort",
+            "values": [
+              "low",
+              "medium",
+              "high",
+              "xhigh"
+            ]
+          }
+        ],
+        "release_date": "2026-05-28",
+        "temperature": true,
+        "tool_call": true
+      },
       "anthropic/claude-sonnet-4": {
         "attachment": true,
         "cost": {
@@ -195637,6 +195663,40 @@
         "temperature": true,
         "tool_call": true
       },
+      "moonshotai/kimi-k3": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.3,
+          "input": 3,
+          "output": 15
+        },
+        "description": "Kimi multimodal agent model for visual understanding, coding, and planning",
+        "family": "kimi-k3",
+        "id": "moonshotai/kimi-k3",
+        "last_updated": "2026-07-16",
+        "limit": {
+          "context": 1000000,
+          "output": 131072
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "pdf"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Kimi K3",
+        "open_weights": false,
+        "reasoning": true,
+        "reasoning_options": [],
+        "release_date": "2026-07-16",
+        "structured_output": true,
+        "temperature": true,
+        "tool_call": true
+      },
       "morph/morph-v3-fast": {
         "attachment": false,
         "cost": {
@@ -197727,6 +197787,31 @@
         "temperature": true,
         "tool_call": false
       },
+      "openai/gpt-realtime-whisper": {
+        "attachment": false,
+        "description": "Streaming speech-to-text model for low-latency transcript deltas from live audio",
+        "family": "whisper",
+        "id": "openai/gpt-realtime-whisper",
+        "last_updated": "2026-05-07",
+        "limit": {
+          "context": 0,
+          "output": 0
+        },
+        "modalities": {
+          "input": [
+            "audio"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "gpt-realtime-whisper",
+        "open_weights": false,
+        "reasoning": false,
+        "release_date": "2026-05-07",
+        "temperature": true,
+        "tool_call": false
+      },
       "openai/o1": {
         "attachment": true,
         "cost": {
@@ -198596,6 +198681,39 @@
         "temperature": true,
         "tool_call": true
       },
+      "thinkingmachines/inkling": {
+        "attachment": true,
+        "cost": {
+          "cache_read": 0.17,
+          "input": 1,
+          "output": 4.05
+        },
+        "description": "Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio",
+        "family": "ling",
+        "id": "thinkingmachines/inkling",
+        "last_updated": "2026-07-15",
+        "limit": {
+          "context": 256000,
+          "output": 256000
+        },
+        "modalities": {
+          "input": [
+            "text",
+            "image",
+            "pdf"
+          ],
+          "output": [
+            "text"
+          ]
+        },
+        "name": "Inkling",
+        "open_weights": true,
+        "reasoning": true,
+        "reasoning_options": [],
+        "release_date": "2026-07-15",
+        "temperature": true,
+        "tool_call": true
+      },
       "voyage/rerank-2.5": {
         "attachment": false,
         "description": "Reranking model for improving retrieval quality in search and recommendation systems",
@@ -199533,8 +199651,7 @@
         },
         "modalities": {
           "input": [
-            "text",
-            "pdf"
+            "text"
           ],
           "output": [
             "text"