MXFP4 Open on Hugging Face
Parameters
27B
Weights (est.)
≈ 16.1 GB
Mac memory (est.)
24 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-vlm on an Apple silicon Mac:
pip install mlx-vlmpython -m mlx_vlm.generate --model mlx-community/Qwopus3.8-27B-Flash-mxfp4-MLX --image photo.jpg --prompt "Describe this image"
Details
- Base model
- Jackrong/Qwopus3.8-27B-Flash (quantized)
- Vendor
- Jackrong
- Task
- Vision + text
- Quantisation
- MXFP4
- Licence
- apache-2.0
- Downloads
- 141 in the last 30 days · 674 all time · 0 likes
- Published
- 2026-09-14 · updated 2026-09-14 (1mo ago)
- Languages
- en, zh, es, ru, ja
- Tags
- qwen3_5, gguf, llama.cpp, vision, multimodal, unsloth, text-generation, qwen3_8, qwen3, qwen, 27b, fine-tuned
Other MLX builds of Jackrong/Qwopus3.8-27B-Flash
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwopus3.8-27B-Flash-mxfp4-MLX (this one) | MXFP4 | ≈ 16.1 GB | 24 GB+ | 141 |
| Qwopus3.8-27B-Flash-mxfp8-MLX | MXFP8 | ≈ 29.5 GB | 48 GB+ | 153 |