Skip to content

Latest commit

 

History

History
91 lines (62 loc) · 2.21 KB

README.md

File metadata and controls

91 lines (62 loc) · 2.21 KB
Raw

llm-watsonx

PyPI License

An IBM watsonx.ai plugin for llm.

Installation

Install this plugin in the same environment as LLM. From the current directory

llm install llm-watsonx

Configuration

You will need to provide the following:

  • API Key from IBM Cloud IAM: https://cloud.ibm.com/iam/apikeys

  • Project ID (from watsonx.ai instance URL: https://dataplatform.cloud.ibm.com/projects//)

  • Associate a Watson Machine Learning (WML) service to your watsonx.ai instance

    1. Inside your watsonx project, navigate to: "Manage" > "Service & Integrations"
    • Alternatively, paste your project ID here and go this URL: https://dataplatform.cloud.ibm.com/projects/<YOUR PROJECT ID>/manage/services?context=wx
    1. Click "Associate service" > Check "WatsonxMachineLearning" > Click "Associate"

    Note: WML service(s) are hosted on your IBM Cloud account (https://cloud.ibm.com/resources). Make sure to have a WML service provisioned and active in your IBM Cloud account for best response rate.

export WATSONX_API_KEY=
export WATSONX_PROJECT_ID=
  • Optionally, if your watsonx instance is not in us-south:
export WATSONX_URL=

Usage

Get list of commands:

llm watsonx --help

Models

See all available models:

llm watsonx list-models

See all generation options:

llm watsonx list-model-options

Example

llm -m watsonx/meta-llama/llama-3-8b-instruct \
    -o temperature .4 \
    -o max_new_tokens 250 \
    "What is IBM watsonx?"

Chat Example

llm chat -m watsonx/meta-llama/llama-3-8b-instruct \
    -o max_new_tokens 1000 \
    -s "You are an assistant for a CLI (command line interface). Provide and help give unix commands to help users achieve their tasks."

Embeddings

See all available models:

llm watsonx list-embedding-models

Example

cat README.md | llm embed -m watsonx/ibm/slate-30m-english-rtrvr