Add WatsonX support #10238
Add WatsonX support #10238
Conversation
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
1 Ignored Deployment
|
dosubot
bot
added
Ɑ: models
|
thank for @baptistebignaud, looking great! last thing is it'd be great to add a demo notebook showing how to use the class to |
|
thanks @baskaryan for reply, I added the notebook |
| def _llm_type(self) -> str: | ||
| return "watsonx" | ||
|
|
||
| def _call( |
There was a problem hiding this comment.
Could you match the method signature exactly:
def _call(
self,
prompt: str,
stop: Optional[List[str]] = None,
run_manager: Optional[CallbackManagerForLLMRun] = None,
**kwargs: Any,
) -> str:
"""Run the LLM on the given prompt and input."""
| class Watsonx(LLM): | ||
| """WatsonX LLM wrapper.""" | ||
|
|
||
| model_name: str = "tiiuae/falcon-40b" |
There was a problem hiding this comment.
Would you be able to document the parameters? Especially ones that might not be obvious for interpretation?
Here's a good example of the format:
| **kwargs: Any, | ||
| ) -> str: | ||
| try: | ||
| import genai |
There was a problem hiding this comment.
Package source: https://ibm.github.io/ibm-generative-ai/index.html
Looks like a name collision with: https://pypi.org/project/genai/ ?
|
Please hold on with that connector since there is a work in progress to use official watsonx.ai library instead of experimental one @baptistebignaud please close the PR. |
|
Great to hear that actions have been officially taken from watsonx.ai; I close the PR |
It is a connector to use a LLM from WatsonX.
It requires python SDK "ibm-generative-ai"
(It might not be perfect since it is my first PR on a public repository 😄)