4-bit Open on Hugging Face
Parameters
49B
Weights (est.)
≈ 27.6 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/JoyAI-LLM-Flash-4bit --prompt "Hello"
Details
- Base model
- jdopensource/JoyAI-LLM-Flash (quantized)
- Vendor
- jdopensource
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- —
- Downloads
- 27 in the last 30 days · 217 all time · 0 likes
- Published
- 2026-02-15 · updated 2026-02-15 (8mo ago)
- Languages
- zh, en
- Tags
- joyai_llm_flash, custom_code
Other MLX builds of jdopensource/JoyAI-LLM-Flash
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| JoyAI-LLM-Flash-4bit-DWQ | 4-bit DWQ | ≈ 27.6 GB | 48 GB+ | 43 |
| JoyAI-LLM-Flash-4bit (this one) | 4-bit | ≈ 27.6 GB | 48 GB+ | 27 |
| JoyAI-LLM-Flash-5bit | 5-bit | ≈ 33.7 GB | 48 GB+ | 23 |
| JoyAI-LLM-Flash-6bit | 6-bit | ≈ 39.8 GB | 48 GB+ | 30 |
| JoyAI-LLM-Flash-8bit | 8-bit | ≈ 52.0 GB | 64 GB+ | 28 |