You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Which version of LM Studio?
Example: LM Studio 0.3.10 (Build 2)
Which operating system?
Windows
What is the bug?
LLMs that use think tags like deepseek-r1-distill-llama-8b@q8_0 sometimes can use [TOOL_REQUEST] blocks within their tags, however it does not seem to be the LLM's intent to call [TOOL_REQUEST] it's just "reasoning" about their input.
This leads to incorrect outputs or repeated calls.
Screenshots
You can see the "red" markings that show the LLM calling [TOOL_REQUEST] blocks.
Blue shows the LM Studio engine calling the tool 3 times.
Yellow shows the result of the LLM straying from the objective.
Logs
Not needed
To Reproduce
Steps to reproduce the behavior:
Have a deepseek-r1-distill-llama-8b@q8_0 model set up with a modestly complex tool.
Have it reason on the response.
You will observe the behavior.
Proposition to solve
LM Studio should ignore TOOL REQUESTS within think tags to avoid this issue.
The text was updated successfully, but these errors were encountered:
Also to add more context for anyone facing the issue, if you are expecting the model to only ever call the tool once, you can always consider the last element in tool call order.
Which version of LM Studio?
Example: LM Studio 0.3.10 (Build 2)
Which operating system?
Windows
What is the bug?
LLMs that use think tags like deepseek-r1-distill-llama-8b@q8_0 sometimes can use [TOOL_REQUEST] blocks within their tags, however it does not seem to be the LLM's intent to call [TOOL_REQUEST] it's just "reasoning" about their input.
This leads to incorrect outputs or repeated calls.
Screenshots
Blue shows the LM Studio engine calling the tool 3 times.
Yellow shows the result of the LLM straying from the objective.
Logs
Not needed
To Reproduce
Steps to reproduce the behavior:
Proposition to solve
LM Studio should ignore TOOL REQUESTS within think tags to avoid this issue.
The text was updated successfully, but these errors were encountered: