| --- |
| license: apache-2.0 |
| base_model: Qwen/Qwen3.6-35B-A3B |
| pipeline_tag: image-text-to-text |
| library_name: mlx |
| tags: |
| - qwen |
| - qwen-3.6 |
| - moe |
| - rotor |
| - mlx |
| - nvfp4 |
| - apple-silicon |
| --- |
| |
| # Qwen3.6-35B-A3B-RotorQuant-MLX-NVFP4 |
|
|
| ## Summary |
|
|
| RotorQuant + MLX-NVFP4 (4-bit) variant of |
| [Qwen/Qwen3.6-35B-A3B](https://huggingface.co/Qwen/Qwen3.6-35B-A3B). |
|
|
| ## Why this variant |
|
|
| Apple Silicon (M1/M2/M3/M4) with RotorQuant structural pre-conditioning and MLX-native NVFP4 layout (E2M1 weights, per-16-element FP8 (NVIDIA Blackwell layout)). 4.503 bits/weight, ~18 GB on disk, sub-2-s load on M4 Max. Pick this over the affine MLX variants when you want NVFP4 format parity with hardware pipelines while running locally. |
|
|
| ## Hardware compatibility |
|
|
| | Device | VRAM | Recommendation | |
| | --- | --- | --- | |
| | Apple M4 Max 128 GB | ~21 GB | recommended — headroom for long context | |
| | Apple M3 Max 64 GB | ~21 GB | fits comfortably | |
| | Apple M2 Max 32 GB | ~21 GB | tight — short context only | |
|
|
| ## Reproduce |
|
|
| ```bash |
| # dequantize from the rotor/turbo MLX-8bit source, then re-quantize |
| python -c "from mlx_lm import convert; convert(hf_path=\"majentik/Qwen3.6-35B-A3B-RotorQuant-MLX-8bit\", mlx_path=\"bf16\", dequantize=True, trust_remote_code=True)" |
| python -c "from mlx_lm import convert; convert(hf_path=\"bf16\", mlx_path=\"out-nvfp4\", quantize=True, q_bits=4, q_group_size=16, q_mode=\"nvfp4\", trust_remote_code=True)" |
| ``` |
|
|
| Reproduced at commit `919836a`. |
|
|
| ## Evaluation |
|
|
| _benchmarks pending — populated after the eval-harness workstream lands._ |
|
|
| ## Family |
|
|
| - **bf16** — [Qwen/Qwen3.6-35B-A3B](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) |
| - **FP8 card** — [majentik/Qwen3.6-35B-A3B-FP8](https://huggingface.co/majentik/Qwen3.6-35B-A3B-FP8) |
| - **RotorQuant MLX-4bit (affine)** — [majentik/Qwen3.6-35B-A3B-RotorQuant-MLX-4bit](https://huggingface.co/majentik/Qwen3.6-35B-A3B-RotorQuant-MLX-4bit) |
| - **RotorQuant MLX-8bit (source for this)** — [majentik/Qwen3.6-35B-A3B-RotorQuant-MLX-8bit](https://huggingface.co/majentik/Qwen3.6-35B-A3B-RotorQuant-MLX-8bit) |
| - **plain MLX-NVFP4 (no rotor/turbo)** — [majentik/Qwen3.6-35B-A3B-MLX-NVFP4](https://huggingface.co/majentik/Qwen3.6-35B-A3B-MLX-NVFP4) |
|
|
| ## Provenance |
|
|
| - Source SHA: `majentik/Qwen3.6-35B-A3B-RotorQuant-MLX-8bit` |
| - Calibration hash: `none (nvfp4 is calibration-free; rotor/turbo conditioning inherited from source)` |
| - Uploaded: `2026-04-21T06:17:30.021158+00:00` |
|
|
| Toolchain: |
| - `huggingface_hub`: 1.11.0 |
| - `mlx`: 0.31.1 |
| - `mlx-lm`: 0.31.2 |
|
|
| ## License |
|
|
| Released under `apache-2.0`. Upstream license of the base model applies. |
|
|
