readable AI benchmarks (simplified)

readable AI benchmarks

Anthropic

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)

reasoning · closed weights · Feb 17, 2026

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) has a Quality Score of 51.2, ranking 83rd 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 56.1, ranking 73rd among 553 scored models, above average (total average: 35.6). Its Value Score is 32.4, around average compared with the total Value average of 30.7.

Quality Score51.2coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score32.4
Factual reliability56.1
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
$2.49

Capability

Intelligence
30.1
Coding
not available
Agentic
not available
Omniscience
12.2
Correct
40.9%
Blended price
$2.31

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 69.5%

Correct 40.9%Incorrect 28.6%Partial / not attempted 30.5%

Artificial Analysis benchmarks

GDPval-AA v2
36%
τ³-Banking
34%
SciCode
50%
Humanity’s Last Exam
34%
GPQA Diamond
87%
CritPt
3%
AA-Omniscience
56%
AA-LCR
80%

Similar models

Support me! Patreon