R1
Open weights
DeepSeek's open-weights reasoning model - the January 2025 release that triggered the "DeepSeek moment" in global AI markets.
- Publisher
- DeepSeek
- Family
- Deepseek R
- Params
- 671B total / 37B active
- Context
- 164K
- Input
- Output
- License
- MIT
- Released
- 20 Jan 2025
Cite
Notes
Only stored in your browser.
Attribution
- Benchmark scores
- AiderAAVaultarena-hard-autoLiveBenchOpenRewardAfterQueryVals AI
Context
164K
Input
$1.68
/1M tok
Output
$4.70
/1M tok
Blended
$2.43
#349/ 414/1M · 3:1
Speed
0
#221/ 414tok/s
Price & speed via Artificial Analysis · ranked across tracked models
Reported on 21 evals across 10 domains - best 96.6% on MATH-500 · #20 of 178
Reported eval scores
21MMLU-Pro84.4
LiveBench66.8
IFBench39
MATH-50096.6
LiveCodeBench61.7
GPQA Diamond70.8
SciCode35.7
AIME202479.8
TaxEval v272.3
CorpFin v254.1