readable AI benchmarks (simplified)

readable AI benchmarks

DeepSeek

DeepSeek R1 (Jan '25)

reasoning · open weights · Jan 20, 2025

DeepSeek R1 (Jan '25) has a Quality Score of 25.7, ranking 299th among 551 scored models. Per 1M tokens, pricing is $2.00 input and $4.00 output. It has a 128k token context window. Its Reliability Score is 33.9, ranking 269th among 553 scored models, below average (total average: 35.6).

Quality Score25.7coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Scorenot available
Factual reliability33.9
Cache Discountnot available

Model specification

Reasoning
reasoning
Input modalities
none
Output modalities
none
Context window
128k tokens
Weights
open weights

Token prices USD per 1M tokens

Input
$2.00
Output
$4.00
Cached Input
not available
Cache write
not available
Cost per task
$0.262

Capability

Intelligence
11.4
Coding
not available
Agentic
not available
Omniscience
-32.2
Correct
30.5%
Blended price
not available

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 93.2%

Correct 30.5%Incorrect 62.7%Partial / not attempted 6.8%

Artificial Analysis benchmarks

GDPval-AA v2
0%
τ³-Banking
6%
SciCode
38%
Humanity’s Last Exam
9%
GPQA Diamond
71%
CritPt
1%
AA-Omniscience
34%
AA-LCR
58%

Similar models

Support me! Patreon