AI projects from Hugging Face

gemma-4-26B-A4B-it Cactus CQ

Cactus-Compute · Kit

2-bit, 3-bit and 4-bit compressed copies of Google's gemma-4-26B-A4B-it chat model in the Cactus weights format. Built to run on ARM devices through the Cactus…

flint-mixed-3-4bit

mlx-community · Kit

A community MLX build of Recursive's 40B Flint coding model that shrinks the experts to 3 bits, so it fits a 24 GB or 32 GB Apple Silicon Mac. Made for develop…

Qwen3.8-27B-GSQ-RCO-IQ3_S-recovered-GGUF

Joakimpalm-Zen · Kit

Shrinks Qwen3.8-27B to a 3-bit IQ3_S GGUF with retrained scales. A research file for people who study quantization, not a faithful stand-in for the original mo…

granite-4.2-3b-GGUF

bartowski · Kit

Quantized GGUF copies of IBM's granite-4.2-3b text model, in many sizes from full precision down to small 3-bit files. For people who run a small chat model on…

DeepSeek-V4.1-Flash-1017 TrellisMX K2 Mix

brandonmusic · Kit

A compressed copy of DeepSeek-V4.1-Flash-1017 sized for two 96 GB GPUs, with a few experts stored at higher precision. A research build for people who serve la…

Ornith-1.5-35B-A3B-GSQ-RCO-GGUF

pfeifferj · Kit

Shrinks the Ornith-1.5-35B-A3B text model to 3-bit and 3.5-bit GGUF files, about 13 and 15 GB instead of 69 GB, for people who run local models with llama.cpp.

Qwen3.8-27B-GGUF by bartowski

bartowski · Kit

A set of GGUF quantizations of Qwen3.8-27B by bartowski, from Q8_0 down to 3-bit, made with an importance matrix. For people running a 27B vision-language mode…

DeepSeek-V4.1-Flash-EXL3-3bpw-2x-RTX-PRO-6000

diffbot · Kit

Serves a 3-bit compressed copy of DeepSeek-V4.1-Flash-1017 on two 96 GB RTX PRO 6000 Blackwell cards with vLLM, for developers who host a large coding and agen…