readable AI benchmarks (simplified)

readable AI benchmarks

Anthropic

Claude 4.5 Sonnet (Reasoning)

reasoning · closed weights · Sep 29, 2025

Claude 4.5 Sonnet (Reasoning) has a Quality Score of 40.9, ranking 165th among 551 scored models. Per 1M tokens, pricing is $3.00 input, $15.00 output, $0.300 cached input, and $3.75 cache write. It has a 1M token context window. Its Reliability Score is 50.0, ranking 115th among 553 scored models, above average (total average: 35.6). Its Value Score is 25.9, low compared with the total Value average of 30.7.

Quality Score40.9coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score25.9
Factual reliability50.0
Cache Discount90%

Model specification

Reasoning
reasoning
Input modalities
none
Output modalities
none
Context window
1M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$3.00
Output
$15.00
Cached Input
$0.300
Cache write
$3.75
Cost per task
$0.557

Capability

Intelligence
20.7
Coding
not available
Agentic
not available
Omniscience
-0.1
Correct
32.9%
Blended price
$2.31

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 65.8%

Correct 32.9%Incorrect 33.0%Partial / not attempted 34.2%

Artificial Analysis benchmarks

GDPval-AA v2
20%
τ³-Banking
25%
SciCode
46%
Humanity’s Last Exam
18%
GPQA Diamond
83%
CritPt
1%
AA-Omniscience
50%
AA-LCR
72%

Similar models

Support me! Patreon