Skip to content

index.mdx

Latest commit

 

History

History
67 lines (63 loc) · 2.75 KB

index.mdx

File metadata and controls

67 lines (63 loc) · 2.75 KB
title Weights & Biases Documentation
description View the documentation for all Weights & Biases products
sidebarTitle Docs Home
mode wide

import {Banner} from "/snippets/Banner.jsx"; import {HomeWrapper} from "/snippets/home.jsx"; import {ProductCard} from "/snippets/ProductCard.jsx";

Use W&B Models to manage AI model development. Features include training, fine-tuning, reporting, automating hyperparameter sweeps, and utilizing the model registry for versioning and reproducibility.

• Introduction
• Quickstart
• YouTube Tutorial
<ProductCard
    title="W&B Weave"
    iconSrc="icons/cropped-weave.svg"
    href="/weave"
    subtitle="Use AI models in your app"
>   
    Use W&B Weave to manage AI models in your code. Features include tracing, output evaluation, cost estimates, and a hosted inference service and playground for comparing different large language models (LLMs) and settings.
    <br/>
    <br/>
    • <a href="/weave">Introduction</a><br/>
    • <a href="/weave/quickstart">Quickstart</a><br/>
    • <a href="https://www.youtube.com/watch?v=IQcGGNLN3zo">YouTube Demo</a><br/>
</ProductCard>

<ProductCard
    title="W&B Inference"
    iconSrc="icons/cropped-inference.svg"
    href="/inference"
    subtitle="Access foundation models"
>
    Use W&B Inference to access leading open-source foundation models through an OpenAI-compatible API. Features include multiple model options, usage tracking, and integration with Weave for tracing and evaluation.
    <br/>
    <br/>
    • <a href="/inference">Introduction</a><br/>
    • <a href="https://wandb.ai/inference">Try in Playground</a>
</ProductCard>

<ProductCard
    title="W&B Training"
    iconSrc="icons/cropped-training.svg"
    href="/training"
    subtitle="Post-train your models"
>
    Now in public preview, use W&B Training to post-train large language models using serverless reinforcement learning (RL). Features include fully managed GPU infrastructure, integration with ART and RULER, and automatic scaling for multi-turn agentic tasks.
    <br/>
    <br/>
    • <a href="/training">Introduction</a><br/>
    • <a href="/training/prerequisites">Quickstart</a><br/>
</ProductCard>