Which GPU can run DeepSeek R1-Distill 32B?

Model-to-GPU finder — pick a model, get the VRAM math and every tracked card that fits. All numbers computed live from our sourced hardware database; nothing is estimated without a label.

Model

AI model Whisper large-v3 (1.5B) Coqui XTTS-v2 (0.8B) FLUX.1 dev (12B) Stable Diffusion 3.5 Large (8B) Stable Diffusion XL 1.0 (3.5B) DeepSeek V4 & V4-Flash (284B) Llama 3.1 70B (70.6B) Llama 3.3 70B (70.6B) DeepSeek R1-Distill 32B (32.8B) Qwen 3 32B (32.8B) Gemma 3 27B (27.4B) Mistral Small 3.2 24B (24.2B) Llama 3.1 8B (8B) Wan 2.2 (A14B & TI2V-5B) (27B)

Quantization

Q4_K_M Q5_K_M Q6_K Q8_0 FP16

Context length

4,096 16,384 32,768 131,072

DeepSeek R1-Distill 32B's verified context window is 131,072 tokens; larger selections are clamped to it.

Find matching GPUs

Disclosure: CompareAIHardware.com participates in the Amazon Associates program and earns from qualifying purchases made through “Check price” links below. Affiliate relationships do not influence fit rankings.

GPUs that can run DeepSeek R1-Distill 32B at Q4_K_M

Grouped by headroom over the 22.7 GB requirement. Every card links to full specs; “Check price” appears only where the product is verified available on Amazon.

Runs it comfortably (25%+ VRAM headroom)

Minimum — fits with little headroom

These hold the model but leave little room; keep contexts short.

Needs 2+ GPUs (tensor/pipeline parallel)

One card is too small, but a pair at ~90% parallel efficiency covers the requirement. Setup details: multi-GPU guide.

Apple Silicon options for DeepSeek R1-Distill 32B

macOS lets Apple Silicon GPUs address about 75% of unified memory. Speed is bandwidth-bound — compare the GB/s column.

Go deeper

• Full fit page for DeepSeek R1-Distill 32B — every quantization level, pre-built systems, FAQ.
• VRAM calculator — start from a parameter count instead of a named model.
• Best GPUs for LLM inference — tier-by-tier buying guide.

Methodology: how we source and verify specs. Found a wrong number? File a correction.