dolphin-2.9.1-llama-3-70b-4bit
mlx-community/dolphin-2.9.1-llama-3-70b-4bit
Uncategorised · mlx
4-bit Open on Hugging Face
Parameters
71B
Weights (est.)
≈ 39.7 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.
Details
- Base model
- meta-llama/Meta-Llama-3-70B (finetune)
- Vendor
- Meta · Llama
- Task
- Uncategorised
- Quantisation
- 4-bit
- Licence
- llama3
- Downloads
- 38 in the last 30 days · 642 all time · 1 likes
- Published
- 2024-05-22 · updated 2024-05-22 (2y ago)
- Tags
- llama, generated_from_trainer, axolotl
Other MLX builds of meta-llama/Meta-Llama-3-70B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| dolphin-2.9.1-llama-3-70b-2bit | 2-bit | ≈ 22.1 GB | 32 GB+ | 24 |
| dolphin-2.9.1-llama-3-70b-4bit (this one) | 4-bit | ≈ 39.7 GB | 48 GB+ | 38 |
| dolphin-2.9.1-llama-3-70b-8bit | 8-bit | ≈ 75.0 GB | 96 GB+ | 19 |