DeepSeek: R1
deepseek/deepseek-r1
121st smartest of 182ReasoningToolsJSON
Intelligence
20.1
121st of 182
Design Elo
—
Speed
—
tokens/sec
Latency
—
first token
Input price
$0.700
197th cheapest
Context
164K
16K max out
How it compares
Smarter than34%
of all ranked models
Cheaper than41%
of all ranked models
Overview
DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass....
Benchmarks
independent · Artificial Analysis & Design ArenaArtificial Analysis61th percentile
Intelligence Index
20.1
GPQA Diamond
81%
Humanity's Last Exam
15%
SciCode
40%
Tau²-Bench (agentic)
37%
Providers & pricing (2)
| Provider | In $/M | Out $/M | Context | Uptime |
|---|---|---|---|---|
| Novitafp8 | $0.700 | $2.50 | 64K | 100% |
| Azure | $1.49 | $5.94 | 164K | 96.3% |
Specifications
Context window164K
Max output16K
Knowledge cutoffJul 2024
Input modalitiestext
Output modalitiestext
Prompt caching—
Cache read price—
ModeratedNo
Open weightsdeepseek-ai/DeepSeek-R1 ↗
R1 FAQ
How much does R1 cost?
R1 costs $0.700 per million input tokens and $2.50 per million output tokens via OpenRouter, making it 197th cheapest of 332 paid models.
How smart is R1?
R1 scores 20.1 on the Artificial Analysis Intelligence Index, ranking 121st of 182 benchmarked models, with a GPQA Diamond score of 81%.
What is R1's context window?
R1 supports a 164K-token context window and can output up to 16K tokens. It accepts text input.