readable AI benchmarks (simplified)

readable AI benchmarks

Agnes 2.5 Pro Theta has a Quality Score of 34.2, ranking 129th among 239 scored models. Per 1M tokens, pricing is $0.450 input, $0.900 output, and $0.045 cached input. It has a 1M token context window. Its Reliability Score is 44.4, ranking 146th among 246 scored models, below average (total average: 46.4). Its Value Score is 28.7, around average compared with the total Value average of 28.3.

Quality Score34.2coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available

Value Score28.7

Factual reliability44.4

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: $0.450
Output: $0.900
Cached Input: $0.045
Cache write: not available
Cost per task: $2.46

Capability

Intelligence: 24.0
Omniscience (native net): -11.2
Correct: 21.3%
Blended price: $0.212

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 53.9%

Correct 21.3%Incorrect 32.6%Partial / not attempted 46.1%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1: 1496.0 Elo
AutomationBench-AA: 48.8%
GDP.pdf: 14.2%
FrontierCode v2.0: 16.6%
Terminal-Bench v4.0: 10.1%
AA-Omniscience v1.1: -11.2 native net
HLE-Diamond: 16.8%
CritPt v2 Hard: 26.0%
Terminal-Bench-Science v0.1.0: 3.8%

Standalone evaluations

GDPval-AA v2.1: 1507.0 Elo

Historical evaluations

SciCode v1.0.1: 56.0%
AA Long Context Reasoning v1.1: 84.7%
Humanity's Last Exam: 37.2%
CritPt: 15.1%

Similar models

Graph Open on Artificial Analysis