CoolFace
Modelpublic

dawncr0w/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-OptiQ-5bpw-MLX

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face
4likes413downloads
Model Card

Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive OptiQ 5bpw

Source: ericpandev/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-bf16 Original: HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive

Quantization

FieldValue
MethodOptiQ mixed-precision MLX
Target BPW5
Achieved BPW5.14
Candidate bits4, 8
Group size64
Calibrationoptiq (24 samples)
Model size18.42 GiB
Layer allocation54 high-bit / 337 low-bit
Layer policy entries391

Notes

  • —This is a text-only MLX export; vision components are omitted.
  • —Native MTP is disabled for this build.
  • —The original model is multimodal; this repository contains only the language model weights.
  • —Built from the bf16 source snapshot listed above.

Usage

bash
huggingface-cli download dawncr0w/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-OptiQ-5bpw-MLX --local-dir Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-OptiQ-5bpw-MLX
python -m mlx_lm.generate --model Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-OptiQ-5bpw-MLX --prompt "Write a short greeting."