Signal SentrySearchUpdated Nov 20, 05:30 UTCAM Drop
Models · scorecard

Gemini 4 Hydrogen

What the vendor claims, next to what independent boards measured. The two are never mixed in one column.

At a glance

Maker
Google
Price per 1M tokens
$1.5 in · $7.5 out
Context window
1.05M
Input → output
audio, file, image, text, video → text
Added to OpenRouter
Nov 12, 2026

Source: OpenRouter, as of Nov 20, 06:00 UTC. Model page on OpenRouter

What the vendor claims

Quoted word for word from the vendor's own launch post or model card. Claims, not measurements.

Scores the vendor claims, each with its source and the quoted line
BenchmarkClaimedSource
Terminal-Bench 4.1The source says: more than tripling Terminal-Bench 4.1 scores (57% vs. 18.2%)57%Google DeepMind · Nov 12From Drop #083 · Fri, Nov 13, 2026 · Morning
HLE-DiamondThe source says: it leads its class on HLE-Diamond (58.8%)58.8%Google DeepMind · Nov 12From Drop #083 · Fri, Nov 13, 2026 · Morning
Vals Tax Agent BenchThe source says: ranks #1 overall on Vals Tax Agent Bench (52.3%)52.3%Google DeepMind · Nov 12From Drop #083 · Fri, Nov 13, 2026 · Morning
Multimodal Knowledge Agent (MKA)The source says: It outperforms other models on the new Multimodal Knowledge Agent (MKA) benchmark (53.4%)53.4%Google DeepMind · Nov 12From Drop #083 · Fri, Nov 13, 2026 · Morning
Agents’ Last Exam v1.1The source says: and on Agents’ Last Exam v1.1 (71.6%) for end-to-end professional workflows71.6%Google DeepMind · Nov 12From Drop #083 · Fri, Nov 13, 2026 · Morning

Independent results

Measured by independent sources (Epoch AI, LMArena, OpenRouter's benchmark runs, τ²-bench, Vectara), not by the vendor. "Listed as" is the source's own name for the model, so you can check our match.

Independent benchmark results for this model, each with its source and date
BoardResultRank
Epoch AI · GPQA diamondListed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source96.8%Accuracy2
Epoch AI · FrontierMath-Tiers-1-3-v2-PrivateListed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source88.1%Accuracy12
Epoch AI · FrontierMath-Tier-4-v2-PrivateListed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source87.8%Accuracy13
Epoch AI · SimpleQA VerifiedListed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source70.9%Accuracy9
Epoch AI · OTIS Mock AIME 2024-2025Listed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source100.0%Accuracy1
Epoch AI · Chess PuzzlesListed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source60.0%Accuracy12
Epoch AI · Mystery Game PuzzlesListed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source65.0%Accuracy8
Epoch AI · Furniture AssemblyListed as: gemini-4-hydrogen_xhighSource: Epoch AI (CC BY), as of Nov 20, 2026. Source57.5%Accuracy13
LMArena · Text (style control)Listed as: Gemini 4 Hydrogen (High)LMArena (CC BY 4.0), as of Nov 18, 2026. Source1,490Arena score16
LMArena · WebDevListed as: Gemini 4 Hydrogen (High)LMArena (CC BY 4.0), as of Nov 18, 2026. Source1,703Arena score14
LMArena · AgentListed as: Gemini 4 Hydrogen (High)LMArena (CC BY 4.0), as of Nov 18, 2026. Source0.135Score10

Price history

OpenRouter's list price per 1M tokens in each of our snapshots.

  • Output now $7.5
  • Input now $1.5

Unchanged across our 15 snapshots.

Show the numbers
This model's OpenRouter price in each of our snapshots
SnapshotInput $/1MOutput $/1M
Nov 20, 06:04 UTC$1.5$7.5
Nov 19, 16:04 UTC$1.5$7.5
Nov 19, 04:04 UTC$1.5$7.5
Nov 18, 18:04 UTC$1.5$7.5
Nov 18, 04:03 UTC$1.5$7.5
Nov 17, 16:01 UTC$1.5$7.5
Nov 17, 04:04 UTC$1.5$7.5
Nov 16, 19:54 UTC$1.5$7.5
Nov 16, 04:05 UTC$1.5$7.5
Nov 15, 16:04 UTC$1.5$7.5
Nov 15, 04:03 UTC$1.5$7.5
Nov 14, 16:04 UTC$1.5$7.5
Nov 14, 04:05 UTC$1.5$7.5
Nov 13, 16:05 UTC$1.5$7.5
Nov 13, 04:04 UTC$1.5$7.5

Source: OpenRouter, as of Nov 20, 06:04 UTC

In the Drops

Models

Google ships Gemini 4 Hydrogen at $1.50/$7.50, plus an Orchestrator that mixes models

Google DeepMind says Hydrogen, its fast Gemini 4 model, finishes tasks 2.5 times faster than 3.8 Flash and scores 57% on Terminal-Bench 4.1. Its Orchestrator in Antigravity switches between Carbon and Hydrogen in one session.

Why it matters: If you run Gemini 3.8 Flash or pay frontier prices for routine agent steps, test Hydrogen as the executor; Google says Orchestrator neared Carbon's success rate on 24% less quota.

On the Frontier watch →From Drop #083 · Fri, Nov 13, 2026 · Morning