readable AI benchmarks (simplified)

readable AI benchmarks

DeepSeek

DeepSeek V3 0324

non-reasoning · open weights · Mar 25, 2025

DeepSeek V3 0324 has a Quality Score of 22.3, ranking 336th among 551 scored models. Per 1M tokens, pricing is $0.845 input, $1.18 output, and $0.793 cached input. It accepts text input, outputs text, and has a 128k token context window. Its Reliability Score is 29.7, ranking 307th among 553 scored models, below average (total average: 35.6). Its Value Score is 15.9, low compared with the total Value average of 30.7.

Quality Score22.3coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score15.9
Factual reliability29.7
Cache Discount6%

Model specification

Reasoning
non-reasoning
Input modalities
text
Output modalities
text
Context window
128k tokens
Weights
open weights

Token prices USD per 1M tokens

Input
$0.845
Output
$1.18
Cached Input
$0.793
Cache write
not available
Cost per task
$0.105

Capability

Intelligence
9.7
Coding
not available
Agentic
not available
Omniscience
-40.7
Correct
24.3%
Blended price
$0.841

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 89.3%

Correct 24.3%Incorrect 65.0%Partial / not attempted 10.7%

Artificial Analysis benchmarks

GDPval-AA v2
0%
τ³-Banking
5%
SciCode
39%
Humanity’s Last Exam
5%
GPQA Diamond
65%
CritPt
0%
AA-Omniscience
30%
AA-LCR
41%

Similar models

Support me! Patreon