Predibase integrates with popular machine learning tools and frameworks to enhance your workflow.

​

LlamaIndex 🦙

LlamaIndex is a framework for building Retrieval Augmented Generation (RAG) applications. The Predibase integration allows you to retrieve contextual information from a vector database and feed this context into an LLM to generate responses.

• Documentation
• End-to-End RAG Tutorial

​

LangChain 🦜🔗

LangChain is a framework for orchestrating tasks in AI applications. The Predibase integration enables seamless integration of hosted OSS models into your AI-native workflows.

• Documentation
• Launch Blog

​

Weights & Biases ⚖️

Weights & Biases is an experiment tracking and logging platform. The integration allows you to track your fine-tuning jobs in Predibase from your W&B dashboard.

​

Setup

1. Create a Weights & Biases account and get an API key from your account settings
2. Navigate to Settings > My Profile in Predibase
3. Paste your W&B API key under “Weights & Biases API Token”
4. Click “Save Token”

Your fine-tuning jobs will automatically be logged to your W&B dashboard under the project “predibase”. The run name will be generated from the adapter repository name and version tag (e.g., my-adapter-repo/1).

​

Comet ☄️

Comet provides experiment tracking and logging for machine learning workloads. The integration enables tracking of fine-tuning jobs and LLM traces.

​

Setup

1. Create a Comet account and get an API key
2. Navigate to Settings > My Profile in Predibase
3. Paste your Comet API key under “Comet API Token”
4. Click “Save Token”

Your fine-tuning jobs will automatically be logged to your Comet dashboard under the project “predibase”. The run name will be your adapter ID (repo-name/version number).

​

Comet Opik

Comet’s Opik platform allows you to log, view and evaluate your LLM traces during development and production. Using Opik’s “LLM as a Judge” evaluators, you can identify and fix issues in your LLM application.

Try Opik with Predibase and Langchain

​

LiteLLM 🚅

LiteLLM handles load balancing, fallbacks and spend tracking across 100+ LLMs in the OpenAI format. The Predibase integration provides access to base open source LLMs and highly efficient fine-tuned adapters.

• Documentation

​

PortKey.AI

PortKey provides an Observability Suite and AI Gateway for LLMs, enabling reliable, cost-efficient, and fast applications in the OpenAI format. The integration gives you access to both base open source LLMs and efficient fine-tuned adapters.

• Documentation