Gemini 4 Hydrogen vs GPT-6 Luna
Gemini 4 Hydrogen leads GPT-6 Luna on coding, 77.9 to 54.9 on the Vibe Coding Index, and GPT-6 Luna costs 15.0x less per output token ($0.50 vs $7.50 per million). Updated November 14, 2026; 49 benchmarks have a result for both.
Gemini 4 Hydrogen or GPT-6 Luna: which to pick
Pick Gemini 4 Hydrogen for coding: it scores 77.9 on the Vibe Coding Index against 54.9 for GPT-6 Luna, though on 4 benchmarks against 5.
Pick GPT-6 Luna for output tokens 15.0x cheaper ($0.50 vs $7.50 per million) and a larger context window (1.05M against 1.05M tokens).
- Shared results
- 49
- Gemini 4 Hydrogen only
- 1
- GPT-6 Luna only
- 69
- Like-for-like categories
- 7 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
Gemini 4 Hydrogen
Gemini 4 Hydrogen scores 77.9 against 54.9.
4 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
No call
Only GPT-6 Luna has a AA Intelligence Index result, 27.9.
Directional only
Web design
Design Arena: websites
Gemini 4 Hydrogen
Gemini 4 Hydrogen leads, 1320 against 1268.
Same board, both measured
General
Vals Index
Gemini 4 Hydrogen
Gemini 4 Hydrogen leads, 68.5% against 52.3%.
Same board, both measured
Long documents
Listed context window
GPT-6 Luna
GPT-6 Luna takes 1.05M tokens against 1.05M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-6 Luna
Gemini 4 Hydrogen $0.0053, GPT-6 Luna $0.0003 per turn.
List prices
At a glance
- Coding index
- Gemini 4 Hydrogen 77.9 from 4 benchmarks, GPT-6 Luna 54.9 from 5 benchmarks.
- Widest category
- Agentic work, 24.2 points apart on average over 6 shared benchmarks.
- Widest single gap
- Terminal-Bench 4 (Vals AI): Gemini 4 Hydrogen 54.0% against GPT-6 Luna 13.6%, 53.9 points apart on a 0 to 100 scale.
- Output price
- Gemini 4 Hydrogen $7.50 against GPT-6 Luna $0.50 per million output tokens, 15.0x apart.
- Context window
- Gemini 4 Hydrogen takes 1.05M tokens, GPT-6 Luna 1.05M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Gemini 4 Hydrogen | Bars | GPT-6 Luna | Ties |
|---|---|---|---|---|
| Coding | 10 | 0 | ||
| Agentic work | 6 | 0 | ||
| Reasoning | 6 | 0 | ||
| Math | 2 | 0 | ||
| Knowledge | 5 | 0 | ||
| Web design | 5 | 0 | ||
| General | 15 | 0 | ||
| All shared boards | 49 | 0 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index100
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| LiveBench coding | 78.5% | 70.8% | Gemini 4 Hydrogen |
| LMArena WebDev | 1704 | 1577 | Gemini 4 Hydrogen |
| Terminal-Bench 4 (Vals AI) | 54.0% | 13.6% | Gemini 4 Hydrogen |
| Vibe Code Bench (Vals AI) | 87.8% | 81.7% | Gemini 4 Hydrogen |
| Code migration (Vals AI) | Not measured | 42.5% | Only GPT-6 Luna measured |
| ProgramBench (Vals AI) | Not measured | 0.5% | Only GPT-6 Luna measured |
| IOI (Vals AI) | Not measured | 55.6% | Only GPT-6 Luna measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 73.0% | Only GPT-6 Luna measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 37.5% | Only GPT-6 Luna measured |
| CyberBench (Vals AI) | Not measured | 76.3% | Only GPT-6 Luna measured |
| SciCode | Not measured | 54.6% | Only GPT-6 Luna measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 54.0% | 12.6% | Gemini 4 Hydrogen |
| FrontierCode 2.0 | 39.0% | 25.2% | Gemini 4 Hydrogen |
| FrontierCode | Not measured | 42.4% | Only GPT-6 Luna measured |
| MirrorCode | Not measured | 26.7% | Only GPT-6 Luna measured |
| ALE-Bench | Not measured | 1576.9 | Only GPT-6 Luna measured |
| LMArena Text: coding | 1516 | 1431 | Gemini 4 Hydrogen |
| LMArena Text: software and IT | 1513 | 1422 | Gemini 4 Hydrogen |
| LMArena WebDev: React | 1724 | 1574 | Gemini 4 Hydrogen |
| LMArena WebDev: HTML | 1743 | 1595 | Gemini 4 Hydrogen |
| LMArena WebDev: from an image | Not measured | 1566 | Only GPT-6 Luna measured |
| GameBench 2 | Not measured | 55.9 | Only GPT-6 Luna measured |
| Terminal-Bench 4.0 | Not measured | 16.4% | Only GPT-6 Luna measured |
| CursorBench 4.0 | 57.8% | Not measured | Only Gemini 4 Hydrogen measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 27.9 | Only GPT-6 Luna measured |
| Epoch Capabilities Index | Not measured | 156.4 | Only GPT-6 Luna measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | Not measured | 1150 | Only GPT-6 Luna measured |
| Design Arena: agents building full-stack apps | Not measured | 1138 | Only GPT-6 Luna measured |
| Design Arena: agents building mobile apps | Not measured | 1117 | Only GPT-6 Luna measured |
| Design Arena: agents building Android apps | Not measured | 1176 | Only GPT-6 Luna measured |
| Design Arena: spreadsheets | Not measured | 875 | Only GPT-6 Luna measured |
| LiveBench research | 86.8% | 48.5% | Gemini 4 Hydrogen |
| LiveBench browser | 77.1% | 70.3% | Gemini 4 Hydrogen |
| LiveBench long agentic | 61.8% | 24.7% | Gemini 4 Hydrogen |
| SREBench (Vals AI) | Not measured | 2.7% | Only GPT-6 Luna measured |
| Finance Agent (Vals AI) | Not measured | 49.9% | Only GPT-6 Luna measured |
| Tax Agent Bench (Vals AI) | 52.3% | 19.9% | Gemini 4 Hydrogen |
| Legal Research Bench (Vals AI) | 53.8% | 30.3% | Gemini 4 Hydrogen |
| Harvey's Legal Agent Benchmark (Vals AI) | 11.3% | 4.2% | Gemini 4 Hydrogen |
| Excel Modeling Benchmark (Vals AI) | Not measured | 68.5% | Only GPT-6 Luna measured |
| BioMysteryBench (Vals AI) | Not measured | 61.5% | Only GPT-6 Luna measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 4.3% | Only GPT-6 Luna measured |
| Vals Science Index | Not measured | 28.4% | Only GPT-6 Luna measured |
| APEX-Agents | Not measured | 44.3% | Only GPT-6 Luna measured |
| SpireClimb | Not measured | 0.0% | Only GPT-6 Luna measured |
| LMArena Search | Not measured | 1160 | Only GPT-6 Luna measured |
| Neither model has a result on the other 24 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| LiveBench reasoning | 88.6% | 81.8% | Gemini 4 Hydrogen |
| MysteryMechanism (Vals AI) | Not measured | 19.4% | Only GPT-6 Luna measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 38.5% | Only GPT-6 Luna measured |
| HLE-Diamond (Artificial Analysis run) | 58.8% | 22.2% | Gemini 4 Hydrogen |
| CritPt v2, hard subset (Artificial Analysis run) | 66.0% | 42.5% | Gemini 4 Hydrogen |
| GPQA Diamond (Epoch AI run) | Not measured | 90.5% | Only GPT-6 Luna measured |
| ARC-AGI | Not measured | 86.7% | Only GPT-6 Luna measured |
| ARC-AGI-2 | Not measured | 59.3% | Only GPT-6 Luna measured |
| Chess puzzles | Not measured | 36.0% | Only GPT-6 Luna measured |
| Mystery game puzzles | Not measured | 7.0% | Only GPT-6 Luna measured |
| LMArena Text: hard prompts | 1520 | 1407 | Gemini 4 Hydrogen |
| LMArena Text: expert | 1558 | 1437 | Gemini 4 Hydrogen |
| LMArena Text: hard prompts in English | 1514 | 1407 | Gemini 4 Hydrogen |
| LMArena Vision: diagrams | Not measured | 1235 | Only GPT-6 Luna measured |
| LMArena Vision: homework | Not measured | 1252 | Only GPT-6 Luna measured |
| GoBench | Not measured | 1120 | Only GPT-6 Luna measured |
| Neither model has a result on the other 10 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| OTIS Mock AIME | Not measured | 98.9% | Only GPT-6 Luna measured |
| FrontierMath, tiers 1 to 3 | Not measured | 78.9% | Only GPT-6 Luna measured |
| FrontierMath, tier 4 | Not measured | 56.1% | Only GPT-6 Luna measured |
| ProofBench | Not measured | 64.0% | Only GPT-6 Luna measured |
| LMArena Text: math | 1507 | 1411 | Gemini 4 Hydrogen |
| LMArena Text: mathematical work | 1534 | 1425 | Gemini 4 Hydrogen |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index50
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| MedCode (Vals AI) | Not measured | 44.7% | Only GPT-6 Luna measured |
| AA-Omniscience | 46.3 | -8.3 | Gemini 4 Hydrogen |
| SimpleQA Verified | Not measured | 41.4% | Only GPT-6 Luna measured |
| LMArena Text: science | 1519 | 1387 | Gemini 4 Hydrogen |
| LMArena Text: medicine and healthcare | 1488 | 1349 | Gemini 4 Hydrogen |
| LMArena Text: legal and government | 1532 | 1392 | Gemini 4 Hydrogen |
| LMArena Text: business and finance | 1486 | 1382 | Gemini 4 Hydrogen |
| FACTS Benchmark Suite | Not measured | 64.8% | Only GPT-6 Luna measured |
| PRBench Legal | Not measured | 42.9% | Only GPT-6 Luna measured |
| Neither model has a result on the other 12 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index50
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| Design Arena: websites | 1320 | 1268 | Gemini 4 Hydrogen |
| Design Arena: UI components | Not measured | 1291 | Only GPT-6 Luna measured |
| Design Arena: game dev | 1369 | 1253 | Gemini 4 Hydrogen |
| Design Arena: data viz | Not measured | 1208 | Only GPT-6 Luna measured |
| Design Arena: 3D | 1415 | 1194 | Gemini 4 Hydrogen |
| Design Arena: all code categories | 1333 | 1261 | Gemini 4 Hydrogen |
| Design Arena: SVG | Not measured | 1235 | Only GPT-6 Luna measured |
| Design Arena: ASCII art | Not measured | 1161 | Only GPT-6 Luna measured |
| LiveBench design | 77.3% | 76.1% | Gemini 4 Hydrogen |
GeneralLike-for-likePublic boards, shown as published, kept out of the index150
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| Vals Index | 68.5% | 52.3% | Gemini 4 Hydrogen |
| LiveBench data analysis | 83.2% | 73.4% | Gemini 4 Hydrogen |
| Public Benefits Bench (Vals AI) | Not measured | 57.6% | Only GPT-6 Luna measured |
| SAGE (Vals AI) | Not measured | 83.4% | Only GPT-6 Luna measured |
| MedScribe (Vals AI) | Not measured | 83.7% | Only GPT-6 Luna measured |
| Long context reasoning (AA-LCR) | Not measured | 83.3% | Only GPT-6 Luna measured |
| GDPval-AA | Not measured | 1443 | Only GPT-6 Luna measured |
| DTBench | Not measured | 90.1% | Only GPT-6 Luna measured |
| LMCA | Not measured | 44.5% | Only GPT-6 Luna measured |
| Furniture assembly | Not measured | 44.2% | Only GPT-6 Luna measured |
| LMArena Text | 1502 | 1388 | Gemini 4 Hydrogen |
| LMArena Text: instruction following | 1519 | 1403 | Gemini 4 Hydrogen |
| LMArena Text: creative writing | 1509 | 1359 | Gemini 4 Hydrogen |
| LMArena Vision | Not measured | 1211 | Only GPT-6 Luna measured |
| LMArena Text: English | 1499 | 1391 | Gemini 4 Hydrogen |
| LMArena Text: non-English | 1495 | 1380 | Gemini 4 Hydrogen |
| LMArena Text: Chinese | 1631 | 1422 | Gemini 4 Hydrogen |
| LMArena Text: German | Not measured | 1384 | Only GPT-6 Luna measured |
| LMArena Text: Japanese | Not measured | 1366 | Only GPT-6 Luna measured |
| LMArena Text: Korean | Not measured | 1347 | Only GPT-6 Luna measured |
| LMArena Text: French | 1512 | 1405 | Gemini 4 Hydrogen |
| LMArena Text: Spanish | 1494 | 1395 | Gemini 4 Hydrogen |
| LMArena Text: Russian | 1492 | 1388 | Gemini 4 Hydrogen |
| LMArena Text: Polish | Not measured | 1382 | Only GPT-6 Luna measured |
| LMArena Text: multi-turn | 1485 | 1393 | Gemini 4 Hydrogen |
| LMArena Text: longer queries | 1524 | 1403 | Gemini 4 Hydrogen |
| LMArena Text: writing and language | 1500 | 1382 | Gemini 4 Hydrogen |
| LMArena Text: entertainment and media | 1503 | 1359 | Gemini 4 Hydrogen |
| LMArena Vision: English | Not measured | 1204 | Only GPT-6 Luna measured |
| LMArena Vision: Chinese | Not measured | 1274 | Only GPT-6 Luna measured |
| LMArena Vision: reading text | Not measured | 1221 | Only GPT-6 Luna measured |
| LMArena Vision: humor | Not measured | 1168 | Only GPT-6 Luna measured |
| LMArena Vision: creative writing | Not measured | 1190 | Only GPT-6 Luna measured |
| Neither model has a result on the other 16 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Gemini 4 Hydrogen | GPT-6 Luna | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
GPT-6 Luna costs less
Repository review
50K input + 3K output tokens
GPT-6 Luna costs less
Agent loop
200K input + 10K output tokens
GPT-6 Luna costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Google Gemini 4 Hydrogen | OpenAI GPT-6 Luna |
|---|---|---|
| Maker | OpenAI | |
| Weights | No public weights listed | No public weights listed |
| Released | November 12, 2026 | September 22, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $1.50 | $0.10 |
| Output $/Mtok | $7.50 | $0.50 |
| Coding index | 77.9 | 54.9 |
| Independent measurements behind it | 4 | 5 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemini 4 Hydrogen better than GPT-6 Luna for coding?
Pick Gemini 4 Hydrogen for coding: it scores 77.9 on the Vibe Coding Index against 54.9 for GPT-6 Luna, though on 4 benchmarks against 5. Pick GPT-6 Luna for output tokens 15.0x cheaper ($0.50 vs $7.50 per million) and a larger context window (1.05M against 1.05M tokens).
Which is cheaper, Gemini 4 Hydrogen or GPT-6 Luna?
Gemini 4 Hydrogen costs $1.50 per million input tokens and $7.50 per million output tokens; GPT-6 Luna costs $0.10 and $0.50. Cost per solved task is known for neither of them.
Which has the bigger context window, Gemini 4 Hydrogen or GPT-6 Luna?
GPT-6 Luna, with 1.05M tokens against 1.05M.
How many benchmarks is the Gemini 4 Hydrogen vs GPT-6 Luna comparison based on?
49 benchmarks have a result for both models as of November 14, 2026. Gemini 4 Hydrogen's coding index rests on 4 independent benchmarks and GPT-6 Luna's on 5. Where only one model has a result, the page says so instead of naming a winner.