RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?API pricesVRAM calculatorNew

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.

378 GBneeded at 4-bit, 8K
0GPUs/Macs: fully in memory
0GPUs: with CPU offload
nonesmallest Mac

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.

ModelParams4-bit · 8KCheapest GPUSmallest Mac
DeepSeek R1684B MoE377.8 GBnone in VRAMnone
DeepSeek R1 Distill Llama 70B71B43.3 GBnone in VRAMMac M2 Max 96GB
DeepSeek R1 Distill Qwen 32B33B21.2 GBRTX 5090 ($1999)Mac M3 Pro 36GB
DeepSeek R1 Distill Qwen 14B15B10.6 GBArc A770 16GB ($349)Mac M4 Pro 24GB
DeepSeek R1 Distill Llama 8B8.0B6.2 GBArc B580 ($249)Mac M4 16GB
DeepSeek R1 Distill Qwen 7B7.6B5.5 GBArc B580 ($249)Mac M4 16GB
DeepSeek R1 Distill Qwen 1.5B1.8B1.9 GBArc 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.

HardwareMemoryVerdictDetails
RX 76008 GBNodetails
RTX 40608 GBNodetails
RTX 50608 GBNodetails
RTX 3060 Ti8 GBNodetails
RTX 4060 Ti 8GB8 GBNodetails
RTX 30708 GBNodetails
RTX 3080 10GB10 GBNodetails
Arc B58012 GBNodetails
RTX 3060 12GB12 GBNodetails
RX 7700 XT12 GBNodetails
RTX 507012 GBNodetails
RTX 407012 GBNodetails
RTX 4070 Super12 GBNodetails
RTX 4070 Ti12 GBNodetails
RTX 3080 12GB12 GBNodetails
Arc A770 16GB16 GBNodetails
RTX 5060 Ti 16GB16 GBNodetails
RTX 4060 Ti 16GB16 GBNodetails
RX 7800 XT16 GBNodetails
RX 7900 GRE16 GBNodetails
RX 907016 GBNodetails
RX 9070 XT16 GBNodetails
RTX 5070 Ti16 GBNodetails
RTX 4070 Ti Super16 GBNodetails
RTX 4080 Super16 GBNodetails
RTX 508016 GBNodetails
RTX 408016 GBNodetails
RX 7900 XT20 GBNodetails
RX 7900 XTX24 GBNodetails
RTX 309024 GBNodetails
RTX 409024 GBNodetails
RTX 3090 Ti24 GBNodetails
RTX 509032 GBNodetails
DGX Spark128 GBNodetails
Ryzen AI Max+ 395 128GB128 GBNodetails
Mac M1 16GB16 GBNodetails
Mac M4 16GB16 GBNodetails
Mac M2 24GB24 GBNodetails
Mac M3 24GB24 GBNodetails
Mac M4 Pro 24GB24 GBNodetails
Mac M4 32GB32 GBNodetails
Mac M1 Pro 32GB32 GBNodetails
Mac M3 Pro 36GB36 GBNodetails
Mac M4 Pro 48GB48 GBNodetails
Mac M4 Pro 64GB64 GBNodetails
Mac M1 Max 64GB64 GBNodetails
Mac M4 Max 64GB64 GBNodetails
Mac M2 Max 96GB96 GBNodetails
Mac M3 Max 128GB128 GBNodetails
Mac M4 Max 128GB128 GBNodetails
Mac M2 Ultra 192GB192 GBNodetails
Mac M3 Ultra 256GB256 GBNodetails
Mac M3 Ultra 512GB512 GBQ3_K_M only (below 4-bit)details

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.