4-bit DWQ Open on Hugging Face
Parameters
69B
Weights (est.)
≈ 38.6 GB
Mac memory (est.)
48 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-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/LongCat-Flash-Lite-4bit-DWQ --prompt "Hello"
Details
- Base model
- meituan-longcat/LongCat-Flash-Lite (quantized)
- Vendor
- meituan-longcat
- Task
- Text generation
- Quantisation
- 4-bit DWQ
- Licence
- mit
- Downloads
- 33 in the last 30 days · 495 all time · 1 likes
- Published
- 2026-01-30 · updated 2026-01-30 (9mo ago)
- Tags
- longcat_flash_ngram, custom_code
Other MLX builds of meituan-longcat/LongCat-Flash-Lite
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| LongCat-Flash-Lite-4bit-DWQ (this one) | 4-bit DWQ | ≈ 38.6 GB | 48 GB+ | 33 |
| LongCat-Flash-Lite-4bit | 4-bit | ≈ 38.6 GB | 48 GB+ | 91 |
| LongCat-Flash-Lite-6bit | 6-bit | ≈ 55.7 GB | 96 GB+ | 42 |
| LongCat-Flash-Lite-8bit | 8-bit | ≈ 72.8 GB | 96 GB+ | 48 |
| LongCat-Flash-Lite-bf16 | bf16 | ≈ 137 GB | 192 GB+ | 24 |