diff --git a/.github/workflows/opencode-review-dispatch.yml b/.github/workflows/opencode-review-dispatch.yml
index 14fcf266..cdc986cc 100644
--- a/.github/workflows/opencode-review-dispatch.yml
+++ b/.github/workflows/opencode-review-dispatch.yml
@@ -3361,6 +3361,14 @@ jobs:
                   "baseURL": "https://opencode.ai/zen/v1"
                 },
                 "models": {
+                  "deepseek-v4-flash-free": {
+                    "name": "DeepSeek V4 Flash Free",
+                    "tool_call": true,
+                    "limit": {
+                      "context": 256000,
+                      "output": 64000
+                    }
+                  },
                   "north-mini-code-free": {
                     "name": "North Mini Code Free",
                     "tool_call": true,
@@ -3694,9 +3702,11 @@ jobs:
           NPM_CONFIG_IGNORE_SCRIPTS: "true"
           NO_COLOR: "1"
           # High-sensitivity review candidates only. Public repositories first
-          # use OpenCode Zen's anonymous North Mini Code free endpoint. Zen's
-          # free-period privacy terms permit retention, so private repositories
-          # never include this candidate. The pool then falls through to
+          # use OpenCode Zen's anonymous DeepSeek V4 Flash Free endpoint, then
+          # North Mini Code Free as fallback. During their free periods,
+          # DeepSeek data may be used for improvement and North data may be
+          # retained and used for improvement, so private repositories never
+          # include either candidate. The pool then falls through to
           # DeepSeek V3, the direct GPT-5.6 Luna slot, pinned PAID
           # OpenRouter coder models (free-tier candidates hit the shared
           # free-models-per-day cap and hung for the full candidate timeout,
@@ -3707,7 +3717,7 @@ jobs:
           # cost-efficient tier, cheaper than the legacy gpt-5 it replaced
           # ($1/$6 vs $1.25/$10 per 1M tokens) so the org OpenAI budget
           # stretches further between top-ups.
-          OPENCODE_MODEL_CANDIDATES: "${{ needs.validate-pr-metadata.outputs.is_private == 'false' && 'opencode-free/north-mini-code-free ' || '' }}github-models/deepseek/deepseek-v3-0324 openai/gpt-5.6-luna openrouter/deepseek/deepseek-v3.2 openrouter/qwen/qwen3-coder github-models/openai/gpt-4.1 github-models/openai/gpt-5 github-models/openai/gpt-5-chat github-models/openai/o3 github-models/deepseek/deepseek-r1-0528 github-models/deepseek/deepseek-r1"
+          OPENCODE_MODEL_CANDIDATES: "${{ needs.validate-pr-metadata.outputs.is_private == 'false' && 'opencode-free/deepseek-v4-flash-free opencode-free/north-mini-code-free ' || '' }}github-models/deepseek/deepseek-v3-0324 openai/gpt-5.6-luna openrouter/deepseek/deepseek-v3.2 openrouter/qwen/qwen3-coder github-models/openai/gpt-4.1 github-models/openai/gpt-5 github-models/openai/gpt-5-chat github-models/openai/o3 github-models/deepseek/deepseek-r1-0528 github-models/deepseek/deepseek-r1"
           # One attempt per model, then fall through to the next model. Retrying
           # the SAME model 5x let a rate-limited/hung leader consume the whole
           # step, so the pool never reached a healthy fallback model.
diff --git a/tests/test_opencode_agent_contract.py b/tests/test_opencode_agent_contract.py
index b9dbb106..9e8399cc 100644
--- a/tests/test_opencode_agent_contract.py
+++ b/tests/test_opencode_agent_contract.py
@@ -92,11 +92,13 @@ def test_opencode_model_pool_sets_high_effort_for_capable_candidates():
     assert candidates_match is not None
     conditional_public_candidate = (
         "${{ needs.validate-pr-metadata.outputs.is_private == 'false' "
-        "&& 'opencode-free/north-mini-code-free ' || '' }}"
+        "&& 'opencode-free/deepseek-v4-flash-free "
+        "opencode-free/north-mini-code-free ' || '' }}"
     )
     candidates_text = candidates_match.group(1)
     assert candidates_text.startswith(conditional_public_candidate)
     candidates = [
+        "opencode-free/deepseek-v4-flash-free",
         "opencode-free/north-mini-code-free",
         *candidates_text.removeprefix(conditional_public_candidate).split(),
     ]
@@ -115,6 +117,7 @@ def test_opencode_model_pool_sets_high_effort_for_capable_candidates():
 
     assert candidate_pairs
     assert candidate_pairs == [
+        ["opencode-free", "deepseek-v4-flash-free"],
         ["opencode-free", "north-mini-code-free"],
         ["github-models", "deepseek/deepseek-v3-0324"],
         ["openai", "gpt-5.6-luna"],
@@ -135,6 +138,20 @@ def test_opencode_model_pool_sets_high_effort_for_capable_candidates():
     assert set(github_candidate_models).issubset(set(github_models))
     assert '"context": 256000' in workflow
     assert '"output": 64000' in workflow
+    generated_config_match = re.search(
+        r"jq -n '(\{.*?\})' >\"\$\{OPENCODE_REVIEW_WORKDIR\}/opencode\.jsonc\"",
+        workflow,
+        re.DOTALL,
+    )
+    assert generated_config_match is not None
+    generated_config = json.loads(generated_config_match.group(1))
+    free_models = generated_config["provider"]["opencode-free"]["models"]
+    deepseek_model = free_models["deepseek-v4-flash-free"]
+    north_model = free_models["north-mini-code-free"]
+    assert deepseek_model["tool_call"] is True
+    assert "response_format" not in deepseek_model.get("options", {})
+    assert north_model["tool_call"] is True
+    assert "response_format" not in north_model["options"]
     assert github_candidate_models == [
         "deepseek/deepseek-v3-0324",
         "openai/gpt-4.1",
@@ -1212,7 +1229,8 @@ def test_workflow_provisions_sandbox_tool_and_reviewer_agent():
     )
     assert (
         "needs.validate-pr-metadata.outputs.is_private == 'false' && "
-        "'opencode-free/north-mini-code-free ' || ''"
+        "'opencode-free/deepseek-v4-flash-free "
+        "opencode-free/north-mini-code-free ' || ''"
     ) in workflow
     assert (
         "github-models/deepseek/deepseek-v3-0324 "