reka-flash-3-4bit
mlx-community/reka-flash-3-4bit
Uncategorised · mlx
4-bit Open on Hugging Face
Parameters
21B
Weights (est.)
≈ 11.8 GB
Mac memory (est.)
16 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.
Details
- Base model
- RekaAI/reka-flash-3 (quantized)
- Vendor
- RekaAI
- Task
- Uncategorised
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 36 in the last 30 days · 246 all time · 1 likes
- Published
- 2025-03-12 · updated 2025-03-18 (1y ago)
- Tags
- llama
Other MLX builds of RekaAI/reka-flash-3
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| reka-flash-3-4bit (this one) | 4-bit | ≈ 11.8 GB | 16 GB+ | 36 |
| reka-flash-3-mlx-fp16 | fp16 | ≈ 41.8 GB | 64 GB+ | 24 |