readable AI benchmarks (simplified)

readable AI benchmarks

Claude Fable 5.2 (Low, Default Fallback) has a Quality Score of 67.7, ranking 6th among 245 scored models. Per 1M tokens, pricing is $8.00 input, $40.00 output, $0.200 cached input, and $10.00 cache write. It has a 1M token context window. Its Reliability Score is 81.5, ranking 5th among 252 scored models, above average (total average: 47.0). Its Value Score is 38.6, high compared with the total Value average of 28.7.

Quality Score67.7coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available

Value Score38.6

Factual reliability81.5

Cache Discount98%

Model specification

Reasoning: reasoning
Input modalities: none
Output modalities: none
Context window: 1M tokens
Weights: closed weights

Token prices USD per 1M tokens

Input: $8.00
Output: $40.00
Cached Input: $0.200
Cache write: $10.00
Cost per task: $3.10

Capability

Intelligence: 53.9
Omniscience (native net): 62.9
Correct: 76.3%
Blended price: $5.74

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 89.7%

Correct 76.3%Incorrect 13.4%Partial / not attempted 10.3%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1: 1562.7 Elo
AutomationBench-AA: 68.0%
GDP.pdf: 26.8%
FrontierCode v2.0: 38.4%
Terminal-Bench v4.0: 54.5%
AA-Omniscience v1.1: 62.9 native net
HLE-Diamond: 60.8%
CritPt v2 Hard: 60.0%
Terminal-Bench-Science v0.1.0: 53.8%

Standalone evaluations

GDPval-AA v2.1: 1545.7 Elo

Historical evaluations

SciCode v1.0.1: 57.8%
AA Long Context Reasoning v1.1: 80.0%
Humanity's Last Exam: 59.7%
CritPt: 29.4%
FrontierCode: 60.6%

Similar models

Graph Open on Artificial Analysis