readable AI benchmarks (simplified)

readable AI benchmarks

DeepSeek

DeepSeek V3.2 (Reasoning)

reasoning · open weights · Dec 1, 2025

DeepSeek V3.2 (Reasoning) has a Quality Score of 35.9, ranking 207th among 551 scored models. Per 1M tokens, pricing is $0.280 input, $0.420 output, and $0.014 cached input. It has a 128k token context window. Its Reliability Score is 38.8, ranking 230th among 553 scored models, above average (total average: 35.6). Its Value Score is 32.0, around average compared with the total Value average of 30.7.

Quality Score35.9coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score32.0
Factual reliability38.8
Cache Discount95%

Model specification

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

Token prices USD per 1M tokens

Input
$0.280
Output
$0.420
Cached Input
$0.014
Cache write
not available
Cost per task
not available

Capability

Intelligence
21.5
Coding
not available
Agentic
not available
Omniscience
-22.5
Correct
33.0%
Blended price
$0.108

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 88.4%

Correct 33.0%Incorrect 55.5%Partial / not attempted 11.6%

Artificial Analysis benchmarks

GDPval-AA v2
9%
Humanity’s Last Exam
25%
GPQA Diamond
84%
CritPt
3%
AA-Omniscience
39%
AA-LCR
73%

Similar models

Support me! Patreon