readable AI benchmarks (simplified)

readable AI benchmarks

Google

Gemini 4 Hydrogen (Xhigh)

reasoning · closed weights · Nov 12, 2026

Gemini 4 Hydrogen (Xhigh) has a Quality Score of 63.5, ranking 21st among 283 scored models. Per 1M tokens, pricing is $1.50 input, $7.50 output, and $0.075 cached input. It accepts text, image, video, audio, and pdf input, outputs text, and has a 1M token context window. Its Reliability Score is 73.1, ranking 16th among 288 scored models, above average (total average: 48.2). Its Value Score is 43.9, high compared with the total Value average of 29.3.

Quality Score63.5coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score43.9
Factual reliability73.1
Cache Discount95%

Model specification

Reasoning
reasoning
Input modalities
text, image, video, audio, pdf
Output modalities
text
Context window
1M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$1.50
Output
$7.50
Cached Input
$0.075
Cache write
not available
Cost per task
$3.16

Capability

Intelligence
53.9
Omniscience (native net)
46.3
Correct
50.5%
Blended price
$1.10

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 54.7%

Correct 50.5%Incorrect 4.2%Partial / not attempted 45.3%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1
1567.5 Elo
AutomationBench-AA
76.1%
private 657-task split from dataset v1.0.6
GDP.pdf
27.6%
FrontierCode v2.0
39.0%
AA Stirrup · 132 tasks, 3 attempts per task
Terminal-Bench v4.0
53.5%
AA mini-swe-agent standardized runs · 66 tasks, pass@1 averaged over 3 repeats per task
AA-Omniscience v1.1
46.3 native net
HLE-Diamond
58.8%
CritPt v2 Hard
66.0%
AA Stirrup · hard subset, 40 of 70 problems
Terminal-Bench-Science v0.1.0
49.5%

Similar models

Support me! Patreon