Downloads · 30 days
24
14% of all-time downloads
qwen3.6-27b-distilled-mlx-q8
chaddy81/qwen3.6-27b-distilled-mlx-q8
qwen3.6-27b-distilled-mlx-q8 is a text generation model from chaddy81. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as other.
MLX 8-bit (8.5 bpw) quantization of the SFT coding-agent model, for Apple-Silicon serving (e.g. oMLX / mlx-lm). Ready to run as-is (~27 GB).
All-time downloads
167
Public
Parameters
26.9B
28.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors28.6 GB · 100%
Parameter types
How the weights are stored.
U3226.9B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Qwen3.6-27B Coding Agent (SFT, distilled) — MLX 8-bit
MLX 8-bit (8.5 bpw) quantization of the SFT coding-agent model, for Apple-Silicon serving (e.g. oMLX / mlx-lm). Ready to run as-is (~27 GB).
Full-precision bf16 merged weights: chaddy81/qwen3.6-27b-distilled.
LoRA adapter only: chaddy81/qwen3.6-27b-distilled-lora.
Training
bf16 LoRA SFT (r=16, alpha=32), 1 epoch / 710 steps, on 5673 teacher-trajectory records (agentic coding: plan → tool calls → verified patch). Merged into the base, then quantized to MLX 8-bit.
Results vs base
Held-out eval (14 tasks × 3; base = untrained Qwen3.6-27B):
| Metric | Base | SFT |
|---|---|---|
| Hardened pass@1 | 0.67 | 0.75 |
| pass@3 | 0.67 | 1.00 |
| Mean files modified | 11.47 | 1.05 |
| Mean tokens | 2.0M | 1.12M |
The standout: files-modified 11.47 → 1.05 — surgical, minimal fixes instead of codebase-wide sprawl.
Run (mlx-lm)
mlx_lm.generate --model chaddy81/qwen3.6-27b-distilled-mlx-q8 --prompt "..."
License
Derivative of Qwen3.6-27B; governed by the base model's license.
- Task
- Text Generation
- Library
- mlx
- Type
- qwen3_5
- License
- other
- Languages
- en
- Created
- Jul 23, 2026
- Updated
- Jul 23, 2026