Advertise here
#48

Gemini 4 Hydrogen

Google Release: 2026-11-12 Tested on: 2026-11-13 00:31 google/gemini-4-hydrogen::xhigh
~300B total (~15B active)MoEClosedEstimated

Gemini 3.8 Flash still tops Gemini 4 Argon, Carbon and Hydrogen in our tests →

Summary

Gemini 4 Hydrogen scores 9.2 on AI BENCHY and ranks #48. It has 9.0 reliability, a 89.9% pass rate, $8.734 total cost, and 89.92s average response time.

What makes Gemini 4 Hydrogen unique: It stands out most in Trivia, where it ranks #1, while Combined is its weakest area at #15. It uses unusually many reasoning tokens, which can help explain its slower or more expensive runs.

Model facts

Researched on 2026-11-12

Estimated
Parameters
~300B total (~15B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. Google does not disclose parameter counts or architecture. Hydrogen is positioned as the high-efficiency Gemini 4 workhorse (Flash tier); the 300B total / 15B active MoE values are a low-confidence catalog estimate from the Gemini Flash family, not vendor-reported. API-only through Google AI Studio and Vertex. OpenRouter declares mandatory reasoning with low, medium, high and xhigh efforts, medium default. The fallback release date records this onboarding request date.

Score

9.2

Consistency

9.1

Total Output Tokens

1,039,092

Total Input Tokens

626,965

Input Price

$1.500 / 1M

Output Price

$7.500 / 1M

Cache Read Price

$0.075 / 1M

Cache Write Price

$0.084 / 1M

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 4

Attempt pass rate: 89.9%

Flaky tests

3

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

89.92s

Response Time (max): 379.79s

Response Time (total): 2068.10s

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#48 Gemini 4 Hydrogen

xhigh
Cost
$0.329
Time
259.1s
Tokens
43,952 tok

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Category Breakdown

Category Score Consistency Tests Correct
Agentic 10.0 10.0
Anti-AI Tricks 10.0 10.0
Coding 10.0 10.0
Combined 5.2 6.9
Data parsing and extraction 10.0 10.0
Domain specific 7.6 7.2
General Intelligence 10.0 10.0
Instructions following 8.5 6.8
Puzzle Solving 10.0 10.0
Tool Calling 10.0 10.0
Trivia 10.0 10.0

Compared models