LM Studio a local 7B model issue "Cannot truncate prompt with n_keep (15028) >= n_ctx (7424)" #11141

Activity

1. github-actions

   assigned

   rekram1-node

   on Jan 29, 2026
2. github-actions commented on Jan 29, 2026

   

   github-actionsBot

   on Jan 29, 2026 – with GitHub Actions

   Contributor

   More actions

   This issue might be a duplicate of existing issues. Please check:

   • OpenCode v1.0.25 Hangs Indefinitely with LM Studio + Qwen Models Due to Empty tool_calls Array #4255: OpenCode v1.0.25 Hangs Indefinitely with LM Studio + Qwen Models Due to Empty tool_calls Array
   • Function call auto-truncation #8012: Function call auto-truncation (related to context window/prompt truncation)
   • Poor performance with local models #4385: Poor performance with local models
   • Only return tool_call parameters but can not run the tool with local LLM #7486: Only return tool_call parameters but can not run the tool with local LLM
   • LM Studio Failure to refresh models #2047: LM Studio Failure to refresh models

   Feel free to ignore if none of these address your specific case.

   👀React with 👀2Reacted by abdulqadar and kuze6969
3. mgonzalezlopezudc commented on Feb 13, 2026

   

   mgonzalezlopezudc

   on Feb 13, 2026

   Last edited by mgonzalezlopezudc

   More actions

   Same behavior here with Qwen 0.6B running on LMstudio. However, setting in LMstudio the context for the model up to maximum, fixes the issue

   👍React with 👍5Reacted by Thomas Roche, Penar Musaraj, Vasanth Muralidharan, Emmanuel LOUISY-GABRIEL and OneShell
4. KevinGimbel commented on Feb 16, 2026

   

   KevinGimbel

   on Feb 16, 2026

   More actions

   Same with openai/gpt-oss-20b, setting context to max fixed it.
5. Jamescd1980 commented on Feb 28, 2026

   

   Jamescd1980

   on Feb 28, 2026

   More actions

   Increasing context size does not resolve the issue. It delays it. Eventually you have too many token to run anything. I'm having the same problem but I can't find away to adjust N_keep, so it will dump the tokens.
6. max-arias commented on Mar 3, 2026

   

   max-arias

   on Mar 3, 2026

   Last edited by max-arias

   More actions

   Same issue with qwen/qwen3.5-9b running through LM Studio 0.4.6, Opencode 1.2.15

   👍React with 👍12Reacted by CaRLymx, Pavel Zamyatin, ixtdfox, ridwanadhip, jericirenej, Chunkai (CK), Michael K, Wilson Silva, Amr Arnous, OneShell and 2 more
7. elecNation commented on Mar 4, 2026

   

   elecNation

   on Mar 4, 2026

   More actions

   通过LM Studio 0.4.6，OpenClaw 运行的qwen/qwen3.5-9b 有同样的问题

   👍React with 👍4Reacted by Wilson Silva, OneShell, Hsin-You Yeh and suerb
8. wilsonsilva commented on Mar 8, 2026

   

   wilsonsilva

   on Mar 8, 2026

   Last edited by wilsonsilva

   More actions

   I'm getting a similar error Cannot truncate prompt with n_keep (11378) >= n_ctx (4096)" with qwen/qwen3.5-9b and opencode 1.2.21.

   Edit: Seems like this isn't an issue with OpenCode itself, but with the configuration in LM Studio. I fixed this by setting the context length to the maximum the model supports (256k)

   

   🎉React with 🎉2Reacted by Luis Taniça and suerb

   ❤️React with ❤️3Reacted by iKlsR, Luis Taniça and suerb
9. Jamroszczyk commented on Mar 9, 2026

   

   Jamroszczyk

   on Mar 9, 2026

   Last edited by Jamroszczyk

   More actions

   1. Increase the context length in Model Defaults
   2. Reload the model into memory; only then will it be applied.

   Consider your available memory, however, before setting it to max. Some models explode with the max value.

   ❤️React with ❤️1Reacted by Praise Dare
10. Jamescd1980 commented on Mar 9, 2026

    

    Jamescd1980

    on Mar 9, 2026

    More actions

    As I said before. Increasing the context window only delays the inevitable issue Wilson and I are having. Increasing context window and setting it to rolling window only works when using the LM UI. When using LM as a server and a deferent program as the UI, like Open Interpreter. You will get the n_keep error. Reloading the model, relaunching LM, rebooting the pc, deleting and reinstalling LM Studio. None of them work to resolve the issue. It as files all over the place that you have to track down and delete in order to fix it. But that has to be done every time you get the n_keep error. The highest context window I can run is 128k. That is about 3.5-4 hours of work.

    👍React with 👍1Reacted by Wilson Silva
11. codinoob1 commented on Mar 12, 2026

    

    codinoob1

    on Mar 12, 2026

    More actions

    While working with LLAMA 7B in Android Studio, I encountered persistent errors that prevented successful execution. I attempted to adjust various settings, but none of the changes resolved the issue.
    The errors consistently appeared during prompt execution, regardless of configuration changes. Unfortunately, I am currently unable to reproduce the issue, but I want to document it here for visibility and potential troubleshooting.
    Steps Taken

    • Adjusted Android Studio project settings (memory allocation, Gradle configs, etc.)
    • Attempted to lower the prompt size
    • Increased the context length
    • Reloaded the environment and retried execution
      Observed Behavior
    • Errors persisted despite configuration adjustments
    • No successful run achieved with LLAMA 7B in Android Studio
      Expected Behavior
    • Model should execute without persistent errors when prompt size and context length are within supported limits
      Notes
    • Issue occurred specifically with LLAMA 7B model
    • May be related to context length handling or prompt size constraints
    • Currently not reproducible, but documenting for future reference and community input
12. github-actions commented on May 12, 2026

    

    github-actionsBot

    on May 12, 2026 – with GitHub Actions

    Contributor

    More actions

    To stay organized issues are automatically closed after 60 days of no activity. If the issue is still relevant please open a new one.
13. github-actions

    closed this as not plannedon May 12, 2026