Lib.rs

tokenizers

today's most used tokenizers, with a focus on performances and versatility

safetensors

functions to read and write safetensors which aim to be safer than their PyTorch counterpart. The format is 8 bytes which is an unsized int, being the size of a JSON header, the JSON…

fastembed

generating vector embeddings, reranking locally

hf-hub

This crates aims ease the interaction with huggingface It aims to be compatible with the huggingface_hub python package, but only implements a smaller subset of functions

llm-tokenizer

LLM tokenizer library with caching and chat template support

hf-xet

Client library and tooling for the Hugging Face Xet data storage system

xet-data

Data processing pipeline for chunking, deduplication, and file reconstruction; used in the Hugging Face Xet client tools. Intended to be used through the API in the hf-xet package.

xet-core-structures

Core data structures including MerkleHash, metadata shards, and Xorb objects

xet-client

Client library for communicating with Hugging Face Xet storage servers. Use through the hf-xet crate.

hf-fetch-model

Download, inspect, and compare HuggingFace models from Rust. Multi-connection parallel downloads plus safetensors header inspection via HTTP Range. No weight data downloaded.

rust-hf-downloader

TUI and CLI for searching and downloading HuggingFace models

synaptic-openai

OpenAI integration for Synaptic: ChatModel + Embeddings

toktrie_hf_tokenizers

HuggingFace tokenizers library support for toktrie and llguidance

plakat

Local text-to-image, style transfer, LoRA, upscale and color-key CLI built on candle

xet-runtime

Async runtime, configuration, logging, and utility infrastructure for the Hugging Face Xet client tools

puma

A lightweight, high-performance inference engine for local AI

api_huggingface

HuggingFace's API for accessing large language models (LLMs) and embeddings

object_detector

Object detection using ORT and the yoloe-26-seg model. This model can detect multiple objects per image, each having a tag, pixel-level mask, and a boundingbox. It's pretrained, it has a vocabulary of 4000+ objects.

golem-ai-embed-hugging-face

working with Hugging face embeding APIs on Golem Cloud

miktik

A unified, multi-backend tokenizer library for LLMs

hf-kernel-builder

Build Hugging Face Hub kernels

gradio

Client in Rust

synaptic

agent framework with LangChain-compatible architecture

bbpe

Binary byte pair encoding (BPE) trainer and CLI compatible with Hugging Face tokenizers

chat-rs

Build LLM clients with ease, attach them to your tools

hf-mem

CLI to estimate inference memory requirements from the Hugging Face Hub

synaptic-huggingface

HuggingFace Inference API integration for Synaptic: Embeddings

rustyface

A Huggingface downloading CLI tool written in Rust

infernum

CLI - From the depths, intelligence rises

bintensors

high-performance binary tensor serialization format designed to be faster eliminating use of JSON serialization metadata

candle-pipelines

intuitive pipelines for local LLM inference in Rust, powered by Candle. Inspired by Python's Transformers library.

ztensor-cli

Command-line tool for zTensor

voice-tts

Rust TTS library backed by MLX, starting with Kokoro

icebreaker

A local AI chat app powered by 🦀Rust, 🧊iced, 🤗Hugging Face, and 🦙llama.cpp

facecrab

Asset management and model downloading for rusty-genius

model-rs

CLI tool for downloading HuggingFace models and running local LLM inference

pf-registry

ProcessFork registry adapters: file, Hugging Face Hub, S3-compatible, IPFS, local OCI

gradio_macro

A macro for generating Gradio interfaces

flodl-hf

HuggingFace integration for flodl: safetensors, hub, tokenizers, pre-built models

toktrie_hf_downloader

HuggingFace Hub download library support for toktrie and llguidance

synthclaw

Lightweight synthetic data generation library/CLI

llm-kit-huggingface

Hugging Face provider for LLM Kit

autotokenizer

我就只是想要rust能有一個簡單的，從hg上拉下config並製作chat prompt的，也這麼難！要我發明輪子，天啊！

jammi-ai

Embeddable AI engine for inference, embeddings, vector search, and fine-tuning

am-compact

Qwen3 inference engine with Attention Management KV-cache compaction (CUDA + Apple Silicon via MLX)

shimmyjinja

Minimal Jinja-like engine for Hugging Face chat_template strings

blazen-model-cache

Shared model download and cache layer for Blazen local-inference backends

rig-fastembed

Rig vector store index integration for Fastembed. https://github.com/Anush008/fastembed-rs

mutranscriber

transcribe video files using Qwen3-ASR

llmvm-outsource

An llmvm backend which sends text and chat generation requests to known hosted language model providers

whisperforge-convert

Convert HuggingFace Whisper safetensors to Burn with INT8 quantization support

transformers

Looking for HuggingFace Transformers in Rust? Check out candle-pipelines!

tauri-plugin-huggingface-downloader-android

lightweight Tauri 2.0 Android plugin for downloading large files from Hugging Face repositories. Features real-time progress tracking via optimized callbacks, background thread execution…

triplets-hf-source

Hugging Face integration for the triplets data pipeline framework

hmll

Safe, idiomatic Rust bindings to the hmll library for high-performance ML model loading

rten-text

Text tokenization and other ML pre/post-processing functions

voice-stt

Speech-to-text library backed by MLX, starting with Moonshine

paca-cli

Helpers for interacting with llama.cpp

hanzo-engine

Hanzo Engine - canonical inference + embedding engine for the Hanzo stack

kitsune-stt

Speech-to-Text tool using Candle and Voxtral

hugging-face-client

rust implment of Hugging Face Hub API

sonr-daemon

Background daemon for sonr, providing semantic search capabilities using local LLMs

kokorox

Lightning fast text-to-speech with the Kokoro model in rust

smolagents-rs

port of the the HuggingFace smolagents library. Build LLM agents with tools and code execution.

blazen-embed-candle

Local embedding backend for Blazen using HuggingFace candle

tokenizers-enfer

today's most used tokenizers, with a focus on performances and versatility

gradio_client_rs

Async Rust client for Gradio apps

alith-models

Load and Download LLM Models, Metadata, and Tokenizers

dsq-io-huggingface

HuggingFace Hub I/O plugin for dsq

entrenar-shell

Interactive REPL for HuggingFace model exploration and distillation

coze

An egui app for playing with a local open source LLM

burn_dragon_sudoku

Sudoku datasets and training for burn_dragon

training-tools

Training monitoring, HuggingFace integration, and checkpoint management for rust-ai

candle_embed

Text embeddings with Candle. Fast and configurable. Use any model from Hugging Face. CUDA or CPU powered.

candle-hf-hub

crates aims ease the interaction with huggingface It aims to be compatible with huggingface_hub python package…

aleph-alpha-tokenizer

A fast implementation of a wordpiece-inspired tokenizer

oxidized-transformers

Transformers library (not functional yet)

llmvm-outsource-lib

outsource backend for llmvm

canon-embed

Local embedding generation for Canon (MiniLM, CPU-only)

hmll-sys

Low-level FFI bindings to the hmll library for high-performance ML model loading

hf-hub-simple-progress

Simplified download callback for hf-hub

diskard-core

Core library for diskard — developer-aware disk cleanup

content-semantic

UCFP semantic fingerprinting (embedding generation) crate

hf-hub-enfer

crates aims ease the interaction with huggingface It aims to be compatible with huggingface_hub python package…

aiha

🦉 AIHA helps you to find the minimal requirements for any model on the 🤗 Hub

huggingface

huggingface_inference_rs

this package is a small wrapper for hugging face Inference API

spn-native

Native model inference and storage for SuperNovae ecosystem

entrenar-common

Shared infrastructure for entrenar CLI tools

aprender-train-distill

End-to-end knowledge distillation CLI

image_captioner

Create captions for images automatically using the BLIP deep learning model

aprender-train-shell

Interactive REPL for HuggingFace model exploration and distillation

llm-bloom

BLOOM (BigScience Large Open-science Open-access Multilingual Language Model) for the llm ecosystem

token-counter

wc for tokens: count tokens in files with HF Tokenizers

ggml-sys

Raw bindings (i.e. bindgen output) for the ggml library.