Open
Description
What feature would you like to see?
I would like to have native function calling support in ChatAdapter, to better avoid tool calling halucinations in ReAct. This would also allow models to make more than 1 tool call per completion.
Example error:
Adapter ChatAdapter failed to parse the LM response.
LM Response: I need to scrutinize the provided ...
[[ ## next_tool_name ## ]]
[[ ## next_tool_args ## ]]
Expected to find output fields in the LM response: [next_thought, next_tool_name, next_tool_args]
Actual output fields parsed from the LM response: [next_tool_name, next_tool_args]
Would you like to contribute?
- Yes, I'd like to help implement this.
- No, I just want to request it.
Additional Context
LiteLLM supports this here: https://docs.litellm.ai/docs/completion/function_call