Qwen3.8-Flash-Next-Uncensored-oQ5e-mtp
mlx-community/Qwen3.8-Flash-Next-Uncensored-oQ5e-mtp
Vision + text · mlx
5-bit Open on Hugging Face
Parameters
180B
Weights (est.)
≈ 124 GB
Mac memory (est.)
192 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/Qwen3.8-Flash-Next-Uncensored-oQ5e-mtp --image photo.jpg --prompt "Describe this image"
Details
- Base model
- orcarouter/Qwen3.8-Flash-Next-Uncensored (quantized)
- Vendor
- orcarouter
- Task
- Vision + text
- Quantisation
- 5-bit
- Licence
- qwen-community-1.0
- Downloads
- 8.2K in the last 30 days · 14K all time · 35 likes
- Published
- 2026-09-15 · updated 2026-09-16 (1mo ago)
- Languages
- en, zh
- Tags
- qwen4_exp, quantized, mixed-precision, imatrix, qwen3.8, flash-next, vision-language, uncensored, abliterated, apple-silicon
Other MLX builds of orcarouter/Qwen3.8-Flash-Next-Uncensored
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen3.8-Flash-Next-Uncensored-oQ5e-mtp (this one) | 5-bit | ≈ 124 GB | 192 GB+ | 8.2K |
| Qwen3.8-Flash-Next-Uncensored-oQ6e-mtp | 6-bit | ≈ 147 GB | 192 GB+ | 2.0K |