readable AI benchmarks (simplified)

readable AI benchmarks

GPT-6 Luna (Max) has a Quality Score of 36.9, ranking 112th among 237 scored models. Per 1M tokens, pricing is $0.100 input, $0.500 output, $0.010 cached input, and $0.125 cache write. It accepts text, image, and pdf input, outputs text, and has a 1M token context window. Its Reliability Score is 45.9, ranking 133rd among 245 scored models, below average (total average: 46.4). Its Value Score is 33.7, high compared with the total Value average of 28.3.

Quality Score36.9coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available

Value Score33.7

Factual reliability45.9

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: $0.100
Output: $0.500
Cached Input: $0.010
Cache write: $0.125
Cost per task: $0.152

Capability

Intelligence: 27.9
Omniscience (native net): -8.3
Correct: 39.6%
Blended price: $0.077

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 87.5%

Correct 39.6%Incorrect 47.9%Partial / not attempted 12.5%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1: 1337.1 Elo
AutomationBench-AA: 53.2%
GDP.pdf: 22.8%
FrontierCode v2.0: 25.2%
Terminal-Bench v4.0: 12.6%
AA-Omniscience v1.1: -8.3 native net
HLE-Diamond: 22.2%
CritPt v2 Hard: 42.5%
Terminal-Bench-Science v0.1.0: 8.6%

Standalone evaluations

GDPval-AA v2.1: 1444.5 Elo

Historical evaluations

SciCode v1.0.1: 54.6%
AA Long Context Reasoning v1.1: 83.3%
Humanity's Last Exam: 38.5%
CritPt: 19.4%
FrontierCode: 41.9%

Similar models

Graph Open on Artificial Analysis