4-bit Open on Hugging Face
Parameters
35B
Weights (est.)
≈ 20.4 GB
Mac memory (est.)
32 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/Nex-N2.5-mini-OptiQ-4bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- nex-agi/Nex-N2.5-mini (quantized)
- Vendor
- nex-agi
- Task
- Vision + text
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 185 in the last 30 days · 1.1K all time · 2 likes
- Published
- 2026-09-12 · updated 2026-09-12 (2mo ago)
- Tags
- qwen3_5_moe, quantized, mixed-precision, 4bit, 8bit, optiq, apple-silicon, text-generation, qwen3.5, agent
Other MLX builds of nex-agi/Nex-N2.5-mini
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Nex-N2.5-mini-oQ4 | 4-bit | ≈ 19.5 GB | 24 GB+ | 68 |
| Nex-N2.5-mini-OptiQ-4bit (this one) | 4-bit | ≈ 20.4 GB | 32 GB+ | 185 |