readable AI benchmarks (simplified)

readable AI benchmarks

DeepSeek

DeepSeek V3.2 Exp (Reasoning)

reasoning · open weights · Sep 29, 2025

DeepSeek V3.2 Exp (Reasoning) has a Quality Score of 30.0, ranking 249th among 551 scored models. Per 1M tokens, pricing is $0.280 input, $0.420 output, and $0.028 cached input. It has a 128k token context window. Its Reliability Score is 34.5, ranking 265th among 553 scored models, below average (total average: 35.6). Its Value Score is 26.6, low compared with the total Value average of 30.7.

Quality Score30.0coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score26.6
Factual reliability34.5
Cache Discount90%

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
$0.028
Cache write
not available
Cost per task
not available

Capability

Intelligence
16.6
Coding
not available
Agentic
not available
Omniscience
-31.0
Correct
27.6%
Blended price
$0.118

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 86.1%

Correct 27.6%Incorrect 58.5%Partial / not attempted 13.9%

Artificial Analysis benchmarks

Humanity’s Last Exam
15%
GPQA Diamond
80%
CritPt
1%
AA-Omniscience
35%
AA-LCR
72%

Similar models

Support me! Patreon