# Best DeepSeek Models (2026)

> All DeepSeek models ranked by benchmark performance.

Canonical page: https://benchlm.ai/best/deepseek-models

Last updated: September 10, 2026

## Rankings

| Rank | Model | Creator | Type | Context | Score |
|------|-------|---------|------|---------|-------|
| 1 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | Proprietary | 1M | 66.38 |
| 2 | [DeepSeek V3.2](/models/deepseek-v3-2) | DeepSeek | Open Weight | 128K | 56.88 |
| 3 | [DeepSeek V3.2 (Thinking)](/models/deepseek-v3-2-thinking) | DeepSeek | Open Weight | 128K | 55.8 |
| 4 | [DeepSeek LLM 2.0](/models/deepseek-llm-2-0) | DeepSeek | Open Weight | 128K | 52.26 |
| 5 | [DeepSeek V3.1](/models/deepseek-v3-1) | DeepSeek | Open Weight | 128K | 50.61 |
| 6 | [DeepSeek-R1](/models/deepseek-r1) | DeepSeek | Open Weight | 128K | 50.21 |
| 7 | [DeepSeek V3.1 (Reasoning)](/models/deepseek-v3-1-reasoning) | DeepSeek | Open Weight | 128K | 48.8 |
| 8 | [DeepSeek Coder 2.0](/models/deepseek-coder-2-0) | DeepSeek | Open Weight | 128K | 48.09 |
| 9 | [DeepSeekMath V2](/models/deepseekmath-v2) | DeepSeek | Open Weight | 128K | 47.71 |
| 10 | [DeepSeek V3](/models/deepseek-v3) | DeepSeek | Open Weight | 128K | 41.49 |
| 11 | [DeepSeek R1 Distill Qwen 32B](/models/deepseek-r1-distill-qwen-32b) | DeepSeek | Open Weight | 128K | 32.45 |

## Key Takeaways

- Top model: [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) with a score of 66.38
- Best open-weight option: [DeepSeek V3.2](/models/deepseek-v3-2) at #2
- Models included: 11

## Compare the Leaders

- [DeepSeek V4 Pro 0813 vs DeepSeek V3.2](/compare/deepseek-v3-2-vs-deepseek-v4-pro-0813)