Basalt v0.11.0 โ Inference, Speech & Quantization Stack
Collection
42+ verified MLX/ANE models (QTIP, PV-Tuning, GSQ-RCO, TTS, OCR) with 1-click launch in Basalt v0.11.0 (Swift 6.4, 84 Compose profiles). โข 43 items โข Updated
How to use goodmai/Qwen3.8-27B-3bit-PVTuned-MLX with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download goodmai/Qwen3.8-27B-3bit-PVTuned-MLX --local-dir Qwen3.8-27B-3bit-PVTuned-MLX
โก ะะฐะฟัััะธัั ะฒ Basalt v0.11.0 ยท bas-alt.vercel.app (macOS 27+ ยท Swift 6.4 ยท Apple Silicon, ะฑะตะท Python):
basalt serve --model goodmai/Qwen3.8-27B-3bit-PVTuned-MLX --rest
3-bit MLX affine checkpoint of Qwen3.8-27B refined with Basalt PV-Tuning against top-64 teacher log-probabilities with monotone held-out KL rollback.
| Metric | Measured Value |
|---|---|
| Method | 3-bit MLX Affine + Basalt PV-Tuning (#410 / #486) |
| Weights Footprint | 11.77 GB |
WikiText-2 16k PPL (4ร4096) |
4.652 (vs 3-bit RTN 4.702; 4-bit GPTQ teacher 4.130) |
System One Accuracy (N=300) |
293 / 300 (97.7%) |
| Reproduce Command | basalt models convert mlx-community/Qwen3.8-27B-GPTQ-4bit --quantize affine:3:64 --experimental-pv-tune --out ./out |
3-bit
Base model
Qwen/Qwen3.5-27B