GSQ-RCO-GGUF Collection Non-uniform GGUF quantizations via GSQ + RCO: per-tensor mixed precision in standard GGUF form • 5 items • Updated 17 days ago • 111
Unsloth Dynamic 3 Quants Collection Introducing Dynamic v3.1 and v3.0 quants, our new SOTA quantization algorithm - performing the best on Terminal Bench, DeepSWE & more at 512K context. • 11 items • Updated 5 days ago • 124
Qwen3.8 Collection Qwen3.8 models converted to MLX format for Apple silicon. • 12 items • Updated Aug 14 • 26
NVIDIA NVFP4 → MLX Collection NVIDIA post-trained NVFP4 checkpoints (ModelOpt PTQ) transcoded to MLX for Apple Silicon / mlx-node, keeping NVIDIA's per-layer bit allocation. • 1 item • Updated Jul 21 • 2
Ornith-1.0 Collection Ornith-1.0 is a family of open-source LLMs specialized for agentic coding. • 8 items • Updated Sep 20 • 395
Gemma 4 QAT Collection Gemma 4 QAT (Quantization-Aware Training) for 3x less memory use and near original accuracy. • 16 items • Updated 23 days ago • 124
Qwen 3.x MTP Collection MLX MTP drafter checkpoints for Qwen 3.x speculative decoding with mlx-vlm. • 12 items • Updated Jun 1 • 9
Gemma-4-31B-IT-unsloth-mlx Collection Gemma-4-31B-IT (dense vision-language) quantized for Apple Silicon (MLX) — Unsloth Dynamic 2.0 with AWQ imatrix pre-scaling. • 9 items • Updated May 16 • 1
Gemma-4-26B-IT-unsloth-mlx Collection Gemma-4-26B-A4B-IT MoE quantized for Apple Silicon (MLX) — Unsloth Dynamic 2.0 with AWQ imatrix pre-scaling. • 8 items • Updated May 16 • 1
Qwen-3.6-unsloth-mlx Collection AWQ-style pre-scaling using Unsloth's imatrix calibration data, then 3-6-bit affine quantization with the Unsloth mixed-precision recipe via MLX • 18 items • Updated May 15 • 19