readable AI benchmarks (simplified)

readable AI benchmarks

Anthropic

Claude Sonnet 5.5 (Max, Default Fallback)

reasoning · closed weights · Sep 28, 2026

Claude Sonnet 5.5 (Max, Default Fallback) has a Quality Score of 59.9, ranking 34th among 283 scored models. Per 1M tokens, pricing is $2.00 input, $10.00 output, $0.100 cached input, and $2.50 cache write. It accepts text, image, and pdf input, outputs text, and has a 1M token context window. Its Reliability Score is 65.2, ranking 52nd among 288 scored models, above average (total average: 48.2). Its Value Score is 40.0, high compared with the total Value average of 29.3.

Quality Score59.9coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score40.0
Factual reliability65.2
Cache Discount95%

Model specification

Reasoning
reasoning
Input modalities
text, image, pdf
Output modalities
text
Context window
1M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$2.00
Output
$10.00
Cached Input
$0.100
Cache write
$2.50
Cost per task
$10.02

Capability

Intelligence
54.5
Omniscience (native net)
30.5
Correct
51.5%
Blended price
$1.47

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 72.5%

Correct 51.5%Incorrect 21.0%Partial / not attempted 27.5%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1
1815.1 Elo
AutomationBench-AA
71.8%
private 657-task split from dataset v1.0.6
GDP.pdf
25.8%
FrontierCode v2.0
41.7%
AA Stirrup · 132 tasks, 3 attempts per task
Terminal-Bench v4.0
63.6%
AA mini-swe-agent standardized runs · 66 tasks, pass@1 averaged over 3 repeats per task
AA-Omniscience v1.1
30.5 native net
HLE-Diamond
53.3%
CritPt v2 Hard
59.0%
AA Stirrup · hard subset, 40 of 70 problems
Terminal-Bench-Science v0.1.0
53.3%

Standalone evaluations

GDPval-AA v2.1
1837.8 Elo

Historical evaluations

SciCode v1.0.1
61.0%
AA Long Context Reasoning v1.1
82.7%
Humanity's Last Exam
55.0%
CritPt
31.4%
FrontierCode
45.8%

Similar models

Support me! Patreon