Models / DeepSeek
Can an M5 Max (128 GB) run DeepSeek R1 671B?
Full R1 MoE, 671B stored. This is a cluster-sized weight file at every common quant.
- Stored parameters
- 671B
- Active parameters
- 37B
- Q4 total (estimate)
- 407.5 GB
- License
- MIT
Model card. DeepSeek-R1 is the 671B-total MoE (about 37B active). Memory uses the stored total.
Compatibility
| Hardware | Q4 | Q8 | FP16 |
|---|---|---|---|
| M3 Ultra 512GB Apple · 512 GB unified | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run | |
| 2× NVIDIA GB10 NVIDIA · 256 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M3 Ultra 256GB Apple · 256 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M5 Ultra 256GB Apple · 256 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M2 Ultra 192GB Apple · 192 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| NVIDIA GB10 NVIDIA · 128 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| Ryzen AI Max+ 395 128GB AMD · 128 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M1 Ultra 128GB Apple · 128 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M5 Max 128GB Apple · 128 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| 4× RTX 3090 NVIDIA · 96 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX Pro 6000 Blackwell NVIDIA · 96 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M5 Ultra 96GB Apple · 96 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| Ryzen AI Max+ 395 64GB AMD · 64 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| 2× RTX 3090 NVIDIA · 48 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M4 Pro 48GB Apple · 48 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M4 Max 36GB Apple · 36 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M5 Max 36GB Apple · 36 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 5090 NVIDIA · 32 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| Radeon AI Pro R9700 AMD · 32 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| Intel Arc Pro B70 Intel · 32 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M4 32GB Apple · 32 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M6 32GB Apple · 32 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 3090 NVIDIA · 24 GB VRAM
| 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 4090 NVIDIA · 24 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| TITAN RTX NVIDIA · 24 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| Radeon RX 7900 XTX AMD · 24 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M4 Pro 24GB Apple · 24 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M3 Pro 18GB Apple · 18 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 4060 Ti 16GB NVIDIA · 16 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 5060 Ti 16GB NVIDIA · 16 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 5070 Ti NVIDIA · 16 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| M4 16GB Apple · 16 GB unified | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 3060 12GB NVIDIA · 12 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 3080 Ti NVIDIA · 12 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 4070 Ti NVIDIA · 12 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
| RTX 2060 Super NVIDIA · 8 GB VRAM | 408 GB Over estimate No measured run | 714 GB Over estimate No measured run | 1343 GB Over estimate No measured run |
Measured runs
| Hardware | Quant | tok/s | Runtime | Source |
|---|---|---|---|---|
| M3 Ultra 512GB | not stated | 19.89 | llama.cpp | Source · checked · prefill 151 |
| M3 Ultra 512GB | Q4 | 18.43 | MLX | Source |
| RTX 3090 | Q2_K | 8.97 | llama.cpp CUDA | Source |