readable AI benchmarks (simplified)

readable AI benchmarks

Claude Opus 4.8 (Adaptive Reasoning, Max Effort) has a Quality Score of 64.3, ranking 32nd among 551 scored models. Per 1M tokens, pricing is $5.00 input, $25.00 output, $0.500 cached input, and $6.25 cache write. It accepts text, image, and pdf input, outputs text, and has a 1M token context window. Its Reliability Score is 64.4, ranking 27th among 553 scored models, above average (total average: 35.6). Its Value Score is 38.4, high compared with the total Value average of 30.7.

Quality Score64.3coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available

Value Score38.4

Factual reliability64.4

Cache Discount90%

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: $5.00
Output: $25.00
Cached Input: $0.500
Cache write: $6.25
Cost per task: $4.08

Capability

Intelligence: 41.8
Coding: not available
Agentic: not available
Omniscience: 28.8
Correct: 48.8%
Blended price: $3.85

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 68.9%

Correct 48.8%Incorrect 20.1%Partial / not attempted 31.1%

Artificial Analysis benchmarks

GDPval-AA v2: 47%
τ³-Banking: 34%
SciCode: 54%
Humanity’s Last Exam: 49%
GPQA Diamond: 92%
CritPt: 21%
AA-Omniscience: 64%
AA-LCR: 78%

Similar models

Graph Open on Artificial Analysis