Gemini 4 Carbon vs Gemini 4 Hydrogen

Google against Google, the same lab's consecutive models. Prices, scores and speed below are live.

  • Gemini 4 Hydrogen costs 25% less on blended price ($4.50 against $6.00 per 1M tokens, 3 input to 1 output).
  • Gemini 4 Carbon scores higher on the Artificial Analysis Intelligence Index: 60 against 53.9.
  • On the 24 benchmarks both have scores on, Gemini 4 Carbon leads on 19 and Gemini 4 Hydrogen on 4, with 1 tied.
  • Both read up to 1M tokens of context.
  • Gemini 4 Hydrogen is newer: listed Nov 12, 2026, against Nov 10, 2026.

Side by side

Gemini 4 CarbonGemini 4 Hydrogen
LabGoogleGoogle
ListedNov 10, 2026Nov 12, 2026
Input per 1M tokens$3.00$2.25
Output per 1M tokens$15.00$11.25
Cached input per 1M$0.15$0.11
Blended (3:1) per 1M$6.00$4.50
Context window1M1M
Max output944K944K
Knowledge cutoff--
Intelligence index6053.9
Coding index--
ReasoningYesYes
Reads imagesYesYes
Tool useYesYes
Structured outputYesYes
Measured output speed22 tokens/s-

Every benchmark both are scored on

Each model's best score per benchmark, with its source. Scores from different sources or settings are not always comparable; each links to where it was published.

BenchmarkGemini 4 CarbonGemini 4 Hydrogen
AIME100%Epoch AI100%Epoch AI
APEX-Agents84.2%Epoch AI78%Epoch AI
Artificial Analysis Intelligence Index60Artificial Analysis and OpenRouter53.9Artificial Analysis and OpenRouter
Chess Puzzles78%Epoch AI60%Epoch AI
Design Arena: 3D1413 EloDesign Arena1397 EloDesign Arena
Design Arena: Code1335 EloDesign Arena1334 EloDesign Arena
Design Arena: Data Viz1343 EloDesign Arena1287 EloDesign Arena
Design Arena: Game Dev1326 EloDesign Arena1339 EloDesign Arena
Design Arena: UI Component1342 EloDesign Arena1340 EloDesign Arena
Design Arena: Website1305 EloDesign Arena1330 EloDesign Arena
Earthborne Rangers (EBR-bench)87.6Epoch AI70.5Epoch AI
FrontierMath Tier 4 (v2)95.1%Epoch AI87.8%Epoch AI
FrontierMath Tiers 1-3 (v2)90.5%Epoch AI88.1%Epoch AI
GPQA Diamond96.5%Epoch AI96.8%Epoch AI
LMArena (Coding)1564 EloLMArena1512 EloLMArena
LMArena (Overall)1539 EloLMArena1502 EloLMArena
LMArena: Creative Writing1533 EloLMArena1508 EloLMArena
LMArena: Hard Prompts1551 EloLMArena1517 EloLMArena
LMArena: Longer Query1549 EloLMArena1521 EloLMArena
LMArena: Math1537 EloLMArena1537 EloLMArena
LMArena: Multi-Turn1570 EloLMArena1489 EloLMArena
Mystery Game Puzzles73%Epoch AI65%Epoch AI
SimpleBench92.6%Epoch AI91.7%Epoch AI
SimpleQA Verified77.7%Epoch AI70.9%Epoch AI

More comparisons

Where these figures come from

Prices are GenMagic's own, what the studio and the API charge, recomputed from each provider's price list every few minutes. The indices are the Artificial Analysis indices, and every benchmark score links to its source, re-read daily. Measured speed is GenMagic's own generation records, in aggregate. See Google models, Google models, every LLM's price and our sources.