2025-12-19T19:27:34+03:00 INFO [agent] Received prompt request for session: d2be3b23-7b38-48e7-bee1-bc69cfa1566a
2025-12-19T19:27:34+03:00 INFO [agent::thread] Thread::send called with model: nvidia_nemotron-3-nano-30b-a3b
2025-12-19T19:27:52+03:00 ERROR [agent::thread] Turn execution failed: Error rendering prompt with jinja template: "Cannot apply filter "string" to type: NullValue".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:27:52+03:00 ERROR [agent] Error in model response stream: Error rendering prompt with jinja template: "Cannot apply filter "string" to type: NullValue".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:30:37+03:00 INFO [agent] Received prompt request for session: af8edefd-afa5-4072-8ef9-a6b9c8298eee
2025-12-19T19:30:37+03:00 INFO [agent::thread] Thread::send called with model: nvidia_nemotron-3-nano-30b-a3b
2025-12-19T19:30:48+03:00 ERROR [agent::thread] Turn execution failed: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:30:48+03:00 ERROR [agent] Error in model response stream: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:31:05+03:00 INFO [agent] Received prompt request for session: 619dbbb2-fef5-4f4f-bcb0-1589364452c6
2025-12-19T19:31:05+03:00 INFO [agent::thread] Thread::send called with model: nvidia_nemotron-3-nano-30b-a3b
2025-12-19T19:31:19+03:00 ERROR [agent::thread] Turn execution failed: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:31:19+03:00 ERROR [agent] Error in model response stream: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:32:36+03:00 WARN [terminal::terminal_hyperlinks] Timed out processing path hyperlink regexes after 2ms
2025-12-19T19:32:36+03:00 INFO [terminal::terminal_hyperlinks] 1ms time out specified in `terminal.path_hyperlink_timeout_ms`
2025-12-19T19:35:50+03:00 INFO [agent] Received prompt request for session: d16d0a07-bf41-4f0c-bca3-4efe2fea53f7
2025-12-19T19:35:50+03:00 INFO [agent::thread] Thread::send called with model: nvidia_nemotron-3-nano-30b-a3b
2025-12-19T19:36:04+03:00 ERROR [agent::thread] Turn execution failed: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:36:04+03:00 ERROR [agent] Error in model response stream: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:38:55+03:00 ERROR [agent::thread] Turn execution failed: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:38:55+03:00 ERROR [agent] Error in model response stream: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:40:48+03:00 INFO [agent] Received prompt request for session: 0a7c589a-b5da-43d7-a3fa-7c2d98a4377f
2025-12-19T19:40:48+03:00 INFO [agent::thread] Thread::send called with model: nvidia_nemotron-3-nano-30b-a3b
2025-12-19T19:40:58+03:00 ERROR [agent::thread] Turn execution failed: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:40:58+03:00 ERROR [agent] Error in model response stream: Error rendering prompt with jinja template: "Unknown test: sequence".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:52:19+03:00 INFO [project::lsp_store] language server 14: unregistered workspace/DidChangeWatchedFiles capability with id 9f6cc8b3-f532-44eb-8939-9ae460ff0e58
2025-12-19T19:52:19+03:00 INFO [project::lsp_store] language server 14: unregistered workspace/DidChangeWatchedFiles capability with id 66dd4e61-5d1d-4ae0-8962-386b6bdbd63a
2025-12-19T19:53:57+03:00 INFO [agent] Received prompt request for session: 0b5b3ccc-134f-4eb1-8c0c-5eafc73e770e
2025-12-19T19:53:57+03:00 INFO [agent::thread] Thread::send called with model: Devstral
2025-12-19T19:54:33+03:00 ERROR [agent::thread] Turn execution failed: Error rendering prompt with jinja template: "Unknown StringValue filter: safe".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
2025-12-19T19:54:33+03:00 ERROR [agent] Error in model response stream: Error rendering prompt with jinja template: "Unknown StringValue filter: safe".
This is usually an issue with the model's prompt template. If you are using a popular model, you can try to search the model under lmstudio-community, which will have fixed prompt templates. If you cannot find one, you are welcome to post this issue to our discord or issue tracker on GitHub. Alternatively, if you know how to write jinja templates, you can override the prompt template in My Models > model settings > Prompt Template.
Reproduction steps
I'm not sure it's a zed bug, but
Nvidia Nemotron NanoandDevstral2 Smallisn't work.I tested in LMstudio itself, Cherry Studio, VSCode(Continue) and this models works nicely. I tried few jinja templates from huggingface, but they doesn't change much, only error type.
UPD:
I also made test with custom AI compatible provider inside built-in agent for both models and works flawlessly with vLLM and Lemonade
Current vs. Expected behavior
Nemotron
Any prompt suitable
Devstral
Prompt with multiple tool calls, reffers to tool call pipeline in model card i assume.

Zed version and system specs
Zed: v0.217.3+stable.105.80433cb239e868271457ac376673a5f75bc4adb1 (Zed)
OS: Linux Wayland fedora 42
Memory: 30.5 GiB
Architecture: x86_64
GPU: AMD Radeon RX 9070 XT (RADV GFX1201) || radv || Mesa 25.3.2
Attach Zed log file
Zed.log
Relevant Zed settings
settings.json
Relevant Keymap
keymap.json
(for AI issues) Model provider details
LM Studio 0.3.36 (Build 1)
If you are using WSL on Windows, what flavor of Linux are you using?
Fedora