Huffing Face

Models. Inhaled.

A very fast mirror of the open-weights models you were going to download anyway. Same files, same checksums, same tools — just closer.

$ export HF_ENDPOINT=https://huffingface.co

That's it. That's the integration. transformers, huggingface_hub, hf — they all just work, only faster. Anything we haven't huffed yet transparently comes from the source.

25 models
3.25 TB huffed
80 huffs served
40 Gbps of exhale
ModelLicenseSizeHuffsStatus
sentence-transformers/all-mpnet-base-v2apache-2.03.8 GB9fully huffed
BAAI/bge-m3mit4.6 GB7fully huffed
BAAI/bge-reranker-v2-m3apache-2.02.3 GB7fully huffed
zai-org/GLM-5.2mit1.51 TB6huffing…
openai/gpt-oss-120bapache-2.0195.8 GB6huffing…
openai/gpt-oss-20bapache-2.041.3 GB6huffing…
sentence-transformers/all-MiniLM-L6-v2apache-2.0976.9 MB6fully huffed
Qwen/Qwen3-32Bapache-2.065.5 GB3fully huffed
mistralai/Mistral-7B-Instruct-v0.3apache-2.029.0 GB3fully huffed
Qwen/Qwen3-8Bapache-2.016.4 GB3fully huffed
deepseek-ai/DeepSeek-R1-Distill-Llama-8Bmit16.1 GB3fully huffed
Qwen/Qwen2.5-7B-Instructapache-2.015.2 GB3fully huffed
Qwen/Qwen3-4Bapache-2.08.1 GB3huffing…
Qwen/Qwen2.5-1.5B-Instructapache-2.03.1 GB3fully huffed
Qwen/Qwen3-0.6Bapache-2.01.5 GB3fully huffed
Qwen/Qwen2.5-0.5B-Instructapache-2.0999.6 MB3fully huffed
deepseek-ai/DeepSeek-R1-Distill-Qwen-7Bmit15.2 GB1fully huffed
google-bert/bert-base-uncasedapache-2.03.5 GB1fully huffed
nomic-ai/nomic-embed-text-v1.5apache-2.02.2 GB1fully huffed
BAAI/bge-base-en-v1.5mit1.3 GB1fully huffed
Qwen/Qwen3-Embedding-0.6Bapache-2.01.2 GB1fully huffed
BAAI/bge-small-en-v1.5mit401.1 MB1fully huffed
zai-org/GLM-5.2-FP8mit755.7 GB0huffing…
nvidia/GLM-5.2-NVFP4mit464.9 GB0huffing…
empero-ai/Qwythos-9B-Claude-Mythos-5-1M-GGUFapache-2.096.2 GB0huffing…

18 of 25 fully local.

Request a model

If its license lets us redistribute it, your request primes the mirror for everyone.