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
Currently, if you use tool in pne.chat. ReAct prompt pattern will turn on default. Now lots of model support tool call, it more simple and fast. Need to add tool support for language model.
Change: when developer run the following example. Use model tool parameter default, rather than startup a Tool Agent.
import pne
def run_code(code: str):
...
response = pne.chat("give me a bubble sort and run it", tools=[run_code])
For LLM intergate litellm, a compatibility scheme is required. For other solutions, you need to determine whether provider is compatible with tool parameters. For details, see the implementation of langchain.
If want to use ToolAgent in pne.chat, using the following code:
importpnedefrun_code(code: str):
...
response=pne.chat("give me a bubble sort and run it", tools=[run_code], enable_agent=True)
The text was updated successfully, but these errors were encountered:
🚀 Feature Request
Currently, if you use tool in pne.chat. ReAct prompt pattern will turn on default. Now lots of model support tool call, it more simple and fast. Need to add tool support for language model.
Change: when developer run the following example. Use model tool parameter default, rather than startup a Tool Agent.
For LLM intergate litellm, a compatibility scheme is required. For other solutions, you need to determine whether provider is compatible with tool parameters. For details, see the implementation of langchain.
If want to use ToolAgent in pne.chat, using the following code:
The text was updated successfully, but these errors were encountered: