DeepSeek R1
DeepSeek · Released Jan 20, 2025
$0.55
Input price / MTok
#8 of 27$2.19
Output price / MTok
#7 of 27128K
Context window
#26 of 2871.5
GPQA Diamond
#17 of 1997.3
MATH-500
#1 of 1Summary
DeepSeek R1 is a open-source language model from DeepSeek, released on Jan 20, 2025.
At $0.55 input and $2.19 output per million tokens, it ranks #8 of 27 for input price, placing it mid-range on input cost.
Its context window of 128K tokens ranks #26 of 28 for context depth in this dataset.
It targets reasoning, coding workloads.
MIT-licensed open weights; sparked the low-cost reasoning model wave in Jan 2025.
Benchmarks
| Benchmark | Score | Max | % | Rank | Source |
|---|---|---|---|---|---|
| GPQA Diamond | 71.5 | 100 | 71.5% | #17/19 | DeepSeek (self-reported in technical rep... |
| MATH-500 | 97.3 | 100 | 97.3% | only entry | DeepSeek (self-reported in technical rep... |
Similar Models
Compare all 6 modelsModels with the most shared benchmark coverage and closest scores to DeepSeek R1.
DeepSeek R1
GPQA Diamond#17 of 19
Graduate-level science reasoningCompare DeepSeek R1 side by side
Pre-filled with 5 similar models