readable AI benchmarks (simplified)

readable AI benchmarks

GPT-6 Luna (High) has a Quality Score of 32.5, ranking 142nd among 238 scored models. Per 1M tokens, pricing is $0.100 input, $0.500 output, $0.010 cached input, and $0.125 cache write. It has a 1M token context window. Its Reliability Score is 43.6, ranking 153rd among 245 scored models, below average (total average: 46.4). Its Value Score is 29.7, around average compared with the total Value average of 28.3.

Quality Score32.5coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available

Value Score29.7

Factual reliability43.6

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: $0.100
Output: $0.500
Cached Input: $0.010
Cache write: $0.125
Cost per task: $0.073

Capability

Intelligence: 21.4
Omniscience (native net): -12.8
Correct: 39.0%
Blended price: $0.077

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 90.8%

Correct 39.0%Incorrect 51.8%Partial / not attempted 9.2%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1: 1186.6 Elo
AutomationBench-AA: 47.8%
GDP.pdf: 18.2%
FrontierCode v2.0: 18.6%
Terminal-Bench v4.0: 4.5%
AA-Omniscience v1.1: -12.8 native net
HLE-Diamond: 15.6%
CritPt v2 Hard: 31.0%
Terminal-Bench-Science v0.1.0: 2.9%

Standalone evaluations

GDPval-AA v2.1: 1352.1 Elo

Historical evaluations

SciCode v1.0.1: 50.3%
AA Long Context Reasoning v1.1: 79.3%
Humanity's Last Exam: 32.9%
CritPt: 15.4%
FrontierCode: 32.0%

Similar models

Graph Open on Artificial Analysis