Can I run Arete-27B on a Mac with 192 GB?
Yes
Its published weights (54.7 GB) fit in the 144 GB a Mac with 192 GB can use, with room for ≈ 256K tokens of context.
hf download NousResearch/Arete-27BFiles and context on Mac with 192 GB
Each file of Arete-27B (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.
| File | Size | On Mac with 192 GB | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| As published | 54.7 GB | Fits | 256K | 256K |
| BF16 | 53.8 GB | Fits | 256K | 256K |
| Q8_0 | 28.7 GB | Fits | 256K | 256K |
| Q6_K_L | 24.7 GB | Fits | 256K | 256K |
| Q6_K | 23.6 GB | Fits | 256K | 256K |
| Q6_K_S | 22.6 GB | Fits | 256K | 256K |
| Q5_K_M | 20.7 GB | Fits | 256K | 256K |
| Q5_K_S | 19.3 GB | Fits | 256K | 256K |
| Q4_K_L | 18.6 GB | Fits | 256K | 256K |
| Q4_1 | 17.6 GB | Fits | 256K | 256K |
| IQ4_NL | 17.2 GB | Fits | 256K | 256K |
| Q4_K_M | 17.2 GB | Fits | 256K | 256K |
| Q4_K_S | 16.1 GB | Fits | 256K | 256K |
| Q4_0 | 16.1 GB | Fits | 256K | 256K |
| IQ4_XS | 15.2 GB | Fits | 256K | 256K |
| IQ3_M | 14.6 GB | Fits | 256K | 256K |
| Q3_K_L | 13.9 GB | Fits | 256K | 256K |
| Q3_K_M | 13.2 GB | Fits | 256K | 256K |
| IQ3_XS | 12.6 GB | Fits | 256K | 256K |
| Q3_K_S | 12.5 GB | Fits | 256K | 256K |
| IQ3_XXS | 12.1 GB | Fits | 256K | 256K |
| Q2_K | 10.6 GB | Fits | 256K | 256K |
| IQ2_M | 10.3 GB | Fits | 256K | 256K |
| IQ2_S | 9.45 GB | Fits | 256K | 256K |
| IQ2_XS | 8.85 GB | Fits | 256K | 256K |
| IQ2_XXS | 8.64 GB | Fits | 256K | 256K |
Arete-27B on other hardware
- NVIDIA H100Yes
- RTX 5090Yes, with a GGUF version
- RTX 4090Yes, with a GGUF version
- RTX 3090Yes, with a GGUF version
- RTX 5060 Ti 16 GBYes, with a GGUF version
- RTX 3060 12 GBYes, with a GGUF version
- RTX 4060No
- Mac with 32 GBYes, with a GGUF version
- Mac with 64 GBYes, with a GGUF version
- Mac with 128 GBYes
- Mac with 192 GBYes