DeepSeek-R1
Classic open reasoning model for transparent math, coding, and deliberate problem solving
ECI STANDING#116 of 266
SCORE138.97136.9–140.47 CI
FRONTIER LAGLaunched 1.1 months behindScores about what o1 did in December 2024.
SCORES LIKE THESE MODELSGPT-5 nanoOpenAI · Aug 2025139.39
Full Epoch Capabilities Index ranking ↗Qwen3-235B-A22BAlibaba · Apr 2025139.35
Qwen3-235B-A22B-Instruct (Jul 2025)Alibaba · Jul 2025138.92
Qwen3-32BAlibaba · Apr 2025138.51
From the lab
Launch announcement · Jan 20, 2025DeepSeek-R1 ReleaseRELEASE DATEJan 20, 2025
WEIGHTSOpen weights
CONTEXT WINDOW128K
MAX OUTPUT32.8K
REASONINGYes
TOOL CALLINGYes
INPUTStext
OUTPUTStext
Specifications from Models.dev. Availability and limits may differ by provider. Benchmark scores come from Epoch AI under CC BY 4.0.