Downloads · 30 days
36
4% of all-time downloads
philipjohnbasile/hy3-family-mini-qwen35b-reap-v1
hy3-family-mini-qwen35b-reap-v1 is a text generation model from philipjohnbasile. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as apache-2.0.
Explore the model guide · All public work
Downloads · 30 days
36
4% of all-time downloads
All-time downloads
916
Public
Parameters
26.6B
15 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors15 GB · 100%
How the weights are stored.
U3226.6B · 100%
From the Hugging Face model README
Explore the model guide · All public work
| This artifact | |
|---|---|
| Purpose | A smaller Qwen Mini variant with 25% of experts removed and no additional post-prune LoRA heal. |
| Runtime | Autoregressive MLX tooling; see the source project for the pruning and generation recipe. |
| Status | Recorded runtime; see the evidence and limits below. |
| Tensor download | 14.97 GB (13.95 GiB) of root .safetensors files, including any root sidecars. This is a file-size total, not peak RAM. |
| Read first | The project recorded 38/46 passes versus 37/46 for its unpruned comparison. A one-case difference is not broad model superiority. |
A REAP-pruned variant of philipjohnbasile/hy3-family-mini-qwen35b-v1: 25% of experts removed (256→192) via saliency-guided REAP pruning (arXiv:2510.13999), with soul/domain-protected experts kept (coding, math, science, security, design, fullstack, gamedev, legacy, music, art, perfumery). No LoRA heal applied — measured to beat the healed variant and match-or-beat the un-pruned base on the project's 46-case eval suite.
18 GB → 14 GB. 58.5 → 82.2 tok/s (+40%).
The standard "prune then heal" recipe (used successfully on this project's own Hy3-architecture models) was tested here with a real prune-only control for the first time — and healing measurably hurt this checkpoint (71.7% vs the prune-only 82.6% pass rate), root-caused to heal-induced verbosity/stopping-calibration drift. Full writeup: data-model-brain what-doesnt-work.md #36.
mlx_lm.chat --model <this-directory>
mlx_lm.server --model <this-directory> --port 8080 # OpenAI /v1
Qwen3.6 quirk: to turn OFF thinking, pass enable_thinking=False as a DIRECT kwarg to apply_chat_template (the nested form and /no_think are ignored) — or --chat-template-args '{"enable_thinking":false}' if serving via this project's own scripts/08_serve_mlx.py.
⚠️ For strict-JSON/structured output, pin repeat_penalty: 1.0 explicitly in the request — see the base model card for the full incident this warning is based on.
hy3-family-mini-qwen35b-v1 (different base than Hy3-295B, English/agent focus, no built-in tool execution).what-doesnt-work.md #38.Source, recipe, receipts: https://github.com/PhilipJohnBasile/hy3-demolition-mlx