Skip to content

DeepSeek-R1

DeepSeekOpen weightsReasoningVerifiedReleased Jan 20, 2025
Compare with MiMo-V2-FlashSource scores ↗
Overall score
41.1
Rank 136 of 222 · 26.8–55.3
Context window
128K
128,000 tokens
Price per 1M tokens
$0.55 in · $2.19 out
Source: BenchLM
Blended price
$0.96
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Knowledge#11466.7
  2. 2Reasoning63.3
  3. 3Instruction following#10345.9
  4. 4Coding27.5
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (8)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Knowledge

GPQA71.5#77 of 117
MMLU-Pro84.0#24 of 67
MMLU90.8#2 of 28
SimpleQA30.1#5 of 10

Math

MATH-50097.3#5 of 20
AIME 202479.8#8 of 19

Coding

SWE-bench Verified49.2#86 of 98

Instruction following

IFEval83.3#36 of 53

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
List price (BenchLM)$0.55$2.19$0.14

DeepSeek's official DeepSeek-R1 release page lists $0.55 input (cache miss) / $2.19 output per million tokens, with cache-hit input at $0.14.

Arena ratings

Human-preference ratings from LMArena. Best-rated variant shown.

Text1373#181 · 18.5K votes

Open weights

From the Hugging Face repository.

Parameters
684B
Licence
mit
Downloads (30 days)
1.3M

Speed

Measured by us through OpenRouter (median of trials).

Not available on OpenRouter, so we cannot measure it yet.

Related models

Similar score