perplexity-ai-r1-1776-6bit
mlx-community/perplexity-ai-r1-1776-6bit
Uncategorised · mlx
6-bit Open on Hugging Face
Parameters
671B
Weights (est.)
≈ 545 GB
Mac memory (est.)
> 512 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
- perplexity-ai/r1-1776 (quantized)
- Vendor
- perplexity-ai
- Task
- Uncategorised
- Quantisation
- 6-bit
- Licence
- mit
- Downloads
- 47 in the last 30 days · 661 all time · 1 likes
- Published
- 2025-02-25 · updated 2025-02-25 (1y ago)
- Tags
- deepseek_v3, custom_code
Other MLX builds of perplexity-ai/r1-1776
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| perplexity-ai-r1-1776-2bit | 2-bit | ≈ 210 GB | 256 GB+ | 43 |
| perplexity-ai-r1-1776-4bit | 4-bit | ≈ 378 GB | 512 GB+ | 93 |
| perplexity-ai-r1-1776-6bit (this one) | 6-bit | ≈ 545 GB | > 512 GB | 47 |
| perplexity-ai-r1-1776-8bit | 8-bit | ≈ 713 GB | > 512 GB | 54 |
| perplexity-ai-r1-1776-bf16 | bf16 | ≈ 1342 GB | > 512 GB | 61 |