failed to send message: POST "http://127.0.0.1:53737/session/ses_7fcdb17a2ffeA A8MIzJJs741C/message": 400 Bad Request {"name":"UnknownError","data":{"message":"TypeErro r: undefined is not an object (evaluating 'U0.tokens.input')"}} #942

Activity

1. rahul-s-bhatt commented on Jul 13, 2025

   

   rahul-s-bhatt

   on Jul 13, 2025

   More actions

   I m getting same issue when trying to hook it up to a local model via LMStudio.

   Tested methods:

   • Deleted the .config/opencode and .cache/opencode and restarted couple of times.

   Below is the opencode.json for the same.
   { "$schema": "https://opencode.ai/config.json", "provider": { "lmstudio": { "npm": "@ai-sdk/openai-compatible", "name": "LM Studio (local)", "options": { "baseURL": "http://127.0.0.1:1234/v1" }, "models": { "lm-studio/phi-3-mini-4k-instruct": { "name": "phi-3-mini-4k-instruct" } } } } }
2. thdxr

   self-assigned this

   on Jul 14, 2025
3. adityaagarw commented on Jul 26, 2025

   

   adityaagarw

   on Jul 26, 2025

   More actions

   I m getting same issue when trying to hook it up to a local model via LMStudio.

   Tested methods:

   * Deleted the .config/opencode and .cache/opencode and restarted couple of times.
   

   Below is the opencode.json for the same. { "$schema": "https://opencode.ai/config.json", "provider": { "lmstudio": { "npm": "@ai-sdk/openai-compatible", "name": "LM Studio (local)", "options": { "baseURL": "http://127.0.0.1:1234/v1" }, "models": { "lm-studio/phi-3-mini-4k-instruct": { "name": "phi-3-mini-4k-instruct" } } } } }

   For local LLMs , it usually means that the input is greater than the context window (Do check LMStudio server logs). The system prompt along with other prompts might be exceeding your context window. Default 4k context window is too low.

   👍React with 👍1Reacted by Henry Black
4. blackhaj commented on Jul 31, 2025

   

   blackhaj

   on Jul 31, 2025

   More actions

   I had the same issue. Jacking up the context length fixed it:

   

   👍React with 👍1Reacted by Austin Treat Emmons
5. blackhaj commented on Jul 31, 2025

   

   blackhaj

   on Jul 31, 2025

   More actions

   That said, I still get other errors. It does input a response before throwing the error at least

   
6. yagil commented on Aug 1, 2025

   

   yagil

   on Aug 1, 2025

   More actions

   That said, I still get other errors. It does input a response before throwing the error at least

   

   Hey! We have fixes to this in the latest beta: https://lmstudio.ai/beta-releases
7. jayair commented on Aug 2, 2025

   

   jayair

   on Aug 2, 2025

   Contributor

   More actions

   nice @yagil

   👍React with 👍1Reacted by Yagil Burowski
8. rekram1-node commented on Dec 28, 2025

   

   rekram1-node

   on Dec 28, 2025

   Contributor

   More actions

   [automated] Closing due to 90+ days of inactivity. Feel free to reopen if you still need this!
9. rekram1-node

   closed this as completedon Dec 28, 2025
10. bussard76

    added a commit that references this issue on May 12, 2026

    improve den first-run flow and remove transient marketing UI (anomaly…

    ...

    298d984