Can I run DeepSeek R1 on my PC?
DeepSeek R1 (684B, 40B active) needs about 378 GB of GPU memory at 4-bit with an 8K context. No single consumer GPU we track runs it fully in VRAM at 4-bit. No Mac we track runs it at 4-bit under the default memory limit.
Every DeepSeek R1 model we track
Cheapest consumer GPU (launch MSRP) and smallest Mac that run each model fully in memory at 4-bit or better with 8K context.
| Model | Params | 4-bit · 8K | Cheapest GPU | Smallest Mac |
|---|---|---|---|---|
| DeepSeek R1 | 684B MoE | 377.8 GB | none in VRAM | none |
| DeepSeek R1 Distill Llama 70B | 71B | 43.3 GB | none in VRAM | Mac M2 Max 96GB |
| DeepSeek R1 Distill Qwen 32B | 33B | 21.2 GB | RTX 5090 ($1999) | Mac M3 Pro 36GB |
| DeepSeek R1 Distill Qwen 14B | 15B | 10.6 GB | Arc A770 16GB ($349) | Mac M4 Pro 24GB |
| DeepSeek R1 Distill Llama 8B | 8.0B | 6.2 GB | Arc B580 ($249) | Mac M4 16GB |
| DeepSeek R1 Distill Qwen 7B | 7.6B | 5.5 GB | Arc B580 ($249) | Mac M4 16GB |
| DeepSeek R1 Distill Qwen 1.5B | 1.8B | 1.9 GB | Arc B580 ($249) | Mac M4 16GB |
DeepSeek R1 on every GPU and Mac
Best quant that fits fully in memory at 8K context with a speed ceiling; otherwise the llama.cpp offload plan (system RAM needed, estimated speed from DDR4-3200 to DDR5-6000). Macs and unified-memory PCs share one pool, so offload adds nothing there. How offload is estimated.
Questions
How much VRAM do I need for DeepSeek R1?
About 378 GB at 4-bit (Q4_K_M) with an 8K context, including KV cache and runtime overhead. Longer contexts need more; the model page has the full table by quant and context.
Why can a mixture-of-experts model run on a smaller card?
Only 40B of its 684B parameters are used per token. llama.cpp's --n-cpu-moe keeps the expert weights of some layers in system RAM while attention and the KV cache stay on the GPU, so each token reads far less from slow RAM than a dense model of the same size would.
Other popular checks: Can I run gpt-oss-120b?Can I run gpt-oss-20b?Can I run Llama 70B?Can I run Qwen3? · Best GPU by budget · Mac vs PC
Model data: Hugging Face Hub API and GGUF file sizes. GPU specs: vendor pages. Speeds are estimates, not benchmarks. Methodology · Report an error.