AMD-Llama-135m-code-bfloat16
mlx-community/AMD-Llama-135m-code-bfloat16
Uncategorised · mlx
Parameters
134M
Weights (est.)
≈ 268 MB
Mac memory (est.)
8 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
- amd/AMD-Llama-135m-code (finetune)
- Vendor
- amd
- Task
- Uncategorised
- Quantisation
- Not stated
- Licence
- apache-2.0
- Downloads
- 10 in the last 30 days · 239 all time · 0 likes
- Published
- 2024-10-03 · updated 2024-10-03 (2y ago)
- Tags
- llama
Other MLX builds of amd/AMD-Llama-135m-code
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| AMD-Llama-135m-code-4bit | 4-bit | ≈ 75 MB | 8 GB+ | 16 |
| AMD-Llama-135m-code-8bit | 8-bit | ≈ 143 MB | 8 GB+ | 14 |
| AMD-Llama-135m-code | — | ≈ 268 MB | 8 GB+ | 16 |
| AMD-Llama-135m-code-float16 | — | ≈ 268 MB | 8 GB+ | 13 |
| AMD-Llama-135m-code-bfloat16 (this one) | — | ≈ 268 MB | 8 GB+ | 10 |
| AMD-Llama-135m-code-float32 | — | ≈ 536 MB | 8 GB+ | 10 |