DeepSeek R1 Models — Hardware Requirements
13 DeepSeek R1 models from DeepSeek and the community, from the smallest that runs in 1.8 GB of VRAM at Q4 up to 684.5B parameters. Every row links to full quantization tables and GPU compatibility.
All DeepSeek R1 Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| DeepSeek R1 Distill Qwen 1.5B | 1.8B | 1.8 GB | 131K | ||
| DeepSeek R1 Distill Qwen 7B | 7.6B | 5.8 GB | 131K | ||
| DeepSeek R1 Distill Llama 8B | 8.0B | 7.3 GB | 131K | ||
| DeepSeek R1 0528 Qwen3 8B | 8.2B | 7.6 GB | 131K | ||
| DeepSeek R1 Distill Qwen 14B | 14.8B | 12.4 GB | 131K | ||
| DeepSeek R1 Distill Qwen 32B | 32.8B | 24.3 GB | 131K | ||
| DeepSeek R1 Distill Llama 70B | 70.6B | 48 GB | 131K | ||
| DeepSeek R1 | 684.5B | 412.1 GB | 164K | ||
| DeepSeek R1 0528 | 684.5B | 412.2 GB | 164K | ||
| DeepSeek R1 Zero | 684.5B | 412.2 GB | 164K |
Community variants and fine-tunes(2)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| DeepSeek R1 Distill Qwen 14B Abliterated v2 | 14.8B | 12.4 GB | 131K | ||
| DeepSeek R1 Distill Qwen 32B Abliterated | 32.8B | 24.3 GB | 131K |
How DeepSeek R1 Compares — llmrun Composite Score
DeepSeek R1 0528 is the highest-scoring DeepSeek R1 model with an llmrun Composite Score of 103.7 — #49 among 106 open models. The top proprietary model, Gemini 4 Carbon, scores 158.6. Click a model to see its full benchmark breakdown.
Gemini 4 Carbon · proprietary158.6
GPT 6 Astra · proprietary156.2
Claude Opus 5.5 · proprietary153.0
GPT 6.1 Sol · proprietary151.7
GPT 6.1 Terra · proprietary150.1
Kimi K3.1 Pro142.8
MiMo V2.6 Pro RL133.3
Kimi K3132.7
MiMo V2.6 Flash RL131.9
GLM 5.3130.1
DeepSeek V4.1 Flash129.4
DeepSeek V4 Pro 0813129.2
DeepSeek R1 0528103.7
DeepSeek R198.9
Frequently Asked Questions
- How much VRAM do I need to run a DeepSeek R1 model?
- The smallest DeepSeek R1 model, DeepSeek R1 Distill Qwen 1.5B, runs from 1.8 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 0.8 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which DeepSeek R1 models can I run on a 16 GB GPU?
- 6 of 12 DeepSeek R1 models fit in 16 GB of VRAM at Q4_K_M, including DeepSeek R1 0528 Qwen3 8B, DeepSeek R1 Distill Qwen 7B, DeepSeek R1 Distill Qwen 1.5B.
- What is the most popular DeepSeek R1 model to run locally?
- DeepSeek R1 0528 Qwen3 8B is the most downloaded DeepSeek R1 model in local-friendly quantized formats. It runs from 7.6 GB of VRAM at Q4_K_M.
- How do DeepSeek R1 models score on benchmarks?
- DeepSeek R1 0528 leads the family with an llmrun Composite Score of 103.7, ranking #49 among 106 open models, while the top proprietary model, Gemini 4 Carbon, scores 158.6. See the comparison chart above for the full standings.