DeepSeek-R1
Classic open reasoning model for transparent math, coding, and deliberate problem solving
- Strongest at
- Coding Top 39% of the past year's models · one benchmark so far
- Context window
- 128K tokens in · 32.8K out
Capability profile
Scored against 91 models released in the past year. 50 is the average model; further out is better.
No results yet: Reading & documents, Visual reasoning, Computer use
Where these scores come from
Each capability combines the published benchmarks listed here, rescaled so 50 is the reference average and 15 points is one standard deviation. A hollow dot rests on a single benchmark.
- Coding Aider polyglot
- Writing EQ-Bench Creative Writing v3, EQ-Bench Longform Writing
- Reasoning SimpleBench, ARC-AGI-2
- Math OTIS Mock AIME 2024-2025
- Science GPQA diamond, WeirdML v2
The DeepSeek R1 0528 line
2 releases in the DeepSeek R1 0528 line.
A newer release is out: DeepSeek R1 0528.
Epoch benchmark boards
Best result on each board it has run. The tick marks the leader.
DeepSeek's past year
9 release days in the past year, about one every 41 days. The next is expected around Nov 28, 2026. Forecasts ↗
Specifications
Specifications and prices from Models.dev; availability and limits may differ by provider. Benchmarks from Epoch AI (CC BY 4.0), LiveBench and the other sources named above.