Qwopus3.8-27B-Coder-Flash-4bit
mlx-community/Qwopus3.8-27B-Coder-Flash-4bit
Vision + text · transformers
4-bit 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-Coder-Flash-4bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- unsloth/Qwen3.8-27B (quantized)
- Vendor
- unsloth
- Task
- Vision + text
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 833 in the last 30 days · 833 all time · 3 likes
- Published
- 2026-11-02 · updated 2026-11-02 (9d ago)
- Languages
- en, zh
- Tags
- qwen3_5, gguf, llama.cpp, vision, multimodal, qwopus3_8, vllm, fine-tuned, coding, code-generation, reasoning, tool-use
Other MLX builds of unsloth/Qwen3.8-27B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwopus3.8-27B-Coder-Flash-4bit (this one) | 4-bit | ≈ 16.1 GB | 24 GB+ | 833 |
| Qwopus3.8-27B-Coder-Flash-6bit | 6-bit | ≈ 22.8 GB | 32 GB+ | 660 |
| Qwopus3.8-27B-Coder-Flash-8bit | 8-bit | ≈ 29.5 GB | 48 GB+ | 923 |