readable AI benchmarks (simplified)

readable AI benchmarks

Step 5 Preview has a Quality Score of 43.4, ranking 73rd among 230 scored models. Per 1M tokens, pricing is $1.00 input, $2.70 output, and $0.100 cached input. It accepts text, image, and video input, outputs text, and has a 1M token context window. Its Reliability Score is 58.2, ranking 67th among 239 scored models, above average (total average: 46.3). Its Value Score is 32.7, high compared with the total Value average of 28.3.

Quality Score43.4coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available

Value Score32.7

Factual reliability58.2

Cache Discount90%

Model specification

Reasoning: reasoning
Input modalities: text, image, video
Output modalities: text
Context window: 1M tokens
Weights: open weights

Token prices USD per 1M tokens

Input: $1.00
Output: $2.70
Cached Input: $0.100
Cache write: not available
Cost per task: $2.35

Capability

Intelligence: 28.5
Omniscience (native net): 16.4
Correct: 36.4%
Blended price: $0.540

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 56.3%

Correct 36.4%Incorrect 19.9%Partial / not attempted 43.7%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1: 1424.5 Elo
AutomationBench-AA: 51.0%
GDP.pdf: 14.8%
FrontierCode v2.0: 12.0%
Terminal-Bench v4.0: 33.3%
AA-Omniscience v1.1: 16.4 native net
HLE-Diamond: 24.3%
CritPt v2 Hard: 32.5%
Terminal-Bench-Science v0.1.0: 2.4%

Standalone evaluations

GDPval-AA v2.1: 1594.0 Elo

Historical evaluations

SciCode v1.0.1: 58.9%
AA Long Context Reasoning v1.1: 88.3%
Humanity's Last Exam: 46.5%
CritPt: 20.9%
FrontierCode: 38.4%

Similar models

Graph Open on Artificial Analysis