modelgrep
D

DeepSeek: R1

deepseek/deepseek-r1

125th smartest of 180ReasoningToolsJSON
Use via OpenRouter ↗
Intelligence
18.8
125th of 180
Design Elo
Speed
70
124th fastest
Latency
1.5s
first token
Input price
$0.700
184th cheapest
Context
164K
16K max out

How it compares

Smarter than31%
of all ranked models
Faster than60%
of all ranked models
Cheaper than38%
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 · via OpenRouter
Artificial Analysis30th percentile
Intelligence Index
18.8
Coding Index
15.9
Agentic Index
3.8
GPQA Diamond
71%
Humanity's Last Exam
9%
SciCode
36%
Tau²-Bench (agentic)
11%

Providers & pricing (2)

ProviderIn $/MOut $/MUptime
Novitafp8$0.700$2.50100%
Azure$1.49$5.9493.7%

Specifications

Context window164K
Max output16K
Knowledge cutoffJul 2024
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price
ModeratedNo

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 184th cheapest of 298 paid models.

How smart is R1?

R1 scores 18.8 on the Artificial Analysis Intelligence Index, ranking 125th of 180 benchmarked models, with a GPQA Diamond score of 71%.

How fast is R1?

R1 generates around 70 tokens per second with 1.5s time-to-first-token (p50), the 124th fastest tracked model.

What is R1's context window?

R1 supports a 164K-token context window and can output up to 16K tokens. It accepts text input.

Compare head-to-head