huggingface/huggingface.js

Health v2 · maintenance only

Adoption not part of the score

2501 stars · 852 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A collection of TypeScript JavaScript libraries for interacting with the Hugging Face ecosystem, including Hub repository management, unified inference across providers, an MCP client, and GGUF parsing. It ships with full TS types and works in Node, Deno, and browsers.

Use cases

• call hugging face inference api from javascript
• upload and download models from the hugging face hub in node
• run chat completions with llama models via inference providers
• generate images from text prompts in a js app
• build an ai agent with tool calling in typescript
• parse a remote gguf model file in the browser
• connect an mcp client to hugging face models

When to choose

• you need typed JavaScript/TypeScript access to the Hugging Face Hub or Inference API
• you want a unified client across multiple inference providers (SambaNova, Replicate, Together, etc.)
• you're building JS apps or browser tools that interact with HF models, datasets, or Spaces

When to avoid

• you need to run ML models locally in the browser (use Transformers.js instead)
• your stack is Python (use huggingface_hub or transformers)
• you need training or fine-tuning capabilities, which this client does not provide

Facets

library · maturity active

http-client llm-inference sdk agent-framework mcp parser templating machine-learning large-language-models artificial-intelligence developer-tools apis browser bun cross-platform huggingface api-client inference-providers model-hub gguf chat-completions text-to-image monorepo nodejs deno typescript

Member repositories

For agents

markdown · JSON · MCP: product_card(name="huggingface/huggingface.js")