GPT-5.6 Luna vs Llama 4 Maverick
Updated October 21, 2026. 67 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick GPT-5.6 Luna for coding: it scores 61.5 on the Vibe Coding Index against 18.1 for Llama 4 Maverick.
Pick Llama 4 Maverick when output tokens cost 1.8x less.
- Shared results
- 67
- GPT-5.6 Luna only
- 49
- Llama 4 Maverick only
- 17
- Like-for-like categories
- 8 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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, four public coding benchmarks
GPT-5.6 Luna
GPT-5.6 Luna scores 61.5 against 18.1.
2 benchmarks behind each score
Intelligence
AA Intelligence Index
GPT-5.6 Luna
GPT-5.6 Luna leads, 37.3 against 9.3.
Same board, both measured
Agentic work
AA Agentic Index
GPT-5.6 Luna
GPT-5.6 Luna leads, 42.1 against 0.6.
Same board, both measured
Reasoning
GPQA Diamond
GPT-5.6 Luna
GPT-5.6 Luna leads, 91.7% against 69.4%.
Same board, both measured
Math
AIME
No call
Only Llama 4 Maverick has a AIME result, 25.2%.
Directional only
Knowledge
MMLU Pro
GPT-5.6 Luna
GPT-5.6 Luna leads, 86.0% against 79.4%.
Same board, both measured
Web design
Design Arena: websites
GPT-5.6 Luna
GPT-5.6 Luna leads, 1256 against 871.
Same board, both measured
Long documents
Listed context window
GPT-5.6 Luna
GPT-5.6 Luna takes 1.05M tokens against 524K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Llama 4 Maverick
GPT-5.6 Luna $0.0008, Llama 4 Maverick $0.0005 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GPT-5.6 Luna
GPT-5.6 Luna $0.04, Llama 4 Maverick $1.491.
Published attempt cost and pass rate
At a glance
- Coding index
- GPT-5.6 Luna 61.5 from 2 benchmarks, Llama 4 Maverick 18.1 from 2 benchmarks.
- Widest category
- Web design, 41.6 points apart on average over 6 shared benchmarks.
- Widest single gap
- ARC-AGI: GPT-5.6 Luna 88.0% against Llama 4 Maverick 4.4%, 83.6 points apart on a 0 to 100 scale.
- Output price
- GPT-5.6 Luna $1.20 against Llama 4 Maverick $0.6525 per million output tokens, 1.8x apart.
- Context window
- GPT-5.6 Luna takes 1.05M tokens, Llama 4 Maverick 524K.
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 | GPT-5.6 Luna | Bars | Llama 4 Maverick | Ties |
|---|---|---|---|---|
| Coding | 6 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 1 | 0 | ||
| Reasoning | 12 | 0 | ||
| Math | 2 | 0 | ||
| Knowledge | 10 | 0 | ||
| Web design | 6 | 0 | ||
| General | 28 | 0 | ||
| All shared boards | 67 | 0 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index60
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| Aider polyglot | Not measured | 15.6% | Only Llama 4 Maverick measured |
| SWE-bench Verified | Not measured | 21.0% | Only Llama 4 Maverick measured |
| LiveBench coding | 73.6% | Not measured | Only GPT-5.6 Luna measured |
| LMArena WebDev | 1518 | Not measured | Only GPT-5.6 Luna measured |
| AA Coding Index | 71.4 | 16.3 | GPT-5.6 Luna |
| LiveCodeBench (Vals AI) | Not measured | 47.3% | Only Llama 4 Maverick measured |
| SWE-bench Verified (Vals AI) | 93.0% | Not measured | Only GPT-5.6 Luna measured |
| Terminal-Bench 4 (Vals AI) | 11.6% | Not measured | Only GPT-5.6 Luna measured |
| Vibe Code Bench (Vals AI) | 77.1% | Not measured | Only GPT-5.6 Luna measured |
| Code migration (Vals AI) | 44.5% | Not measured | Only GPT-5.6 Luna measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only GPT-5.6 Luna measured |
| IOI (Vals AI) | 61.8% | Not measured | Only GPT-5.6 Luna measured |
| Terminal-Bench 2.1 (Vals AI) | 79.0% | Not measured | Only GPT-5.6 Luna measured |
| Vibe Code Bench 1-100 (Vals AI) | 22.6% | Not measured | Only GPT-5.6 Luna measured |
| SkillsBench (Vals AI) | 60.5% | Not measured | Only GPT-5.6 Luna measured |
| CyberBench (Vals AI) | 73.6% | Not measured | Only GPT-5.6 Luna measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 39.7% | Only Llama 4 Maverick measured |
| SciCode | 53.6% | 31.7% | GPT-5.6 Luna |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 6.8% | Only Llama 4 Maverick measured |
| FrontierCode | 39.8% | Not measured | Only GPT-5.6 Luna measured |
| WeirdML | 60.9% | 24.5% | GPT-5.6 Luna |
| ALE-Bench | 1667.4 | 173.0 | GPT-5.6 Luna |
| LMArena Text: coding | 1466 | 1301 | GPT-5.6 Luna |
| LMArena Text: software and IT | 1463 | 1299 | GPT-5.6 Luna |
| LMArena WebDev: React | 1508 | Not measured | Only GPT-5.6 Luna measured |
| LMArena WebDev: HTML | 1520 | Not measured | Only GPT-5.6 Luna measured |
| LMArena WebDev: from an image | 1517 | Not measured | Only GPT-5.6 Luna measured |
| Neither model has a result on the other 11 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| AA Intelligence Index | 37.3 | 9.3 | GPT-5.6 Luna |
| Epoch Capabilities Index | 156.5 | 132.2 | GPT-5.6 Luna |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| AA Agentic Index | 42.1 | 0.6 | GPT-5.6 Luna |
| LiveBench research | 51.0% | Not measured | Only GPT-5.6 Luna measured |
| LiveBench browser | 68.0% | Not measured | Only GPT-5.6 Luna measured |
| LiveBench long agentic | 30.4% | Not measured | Only GPT-5.6 Luna measured |
| SREBench (Vals AI) | 1.5% | Not measured | Only GPT-5.6 Luna measured |
| Finance Agent (Vals AI) | 55.0% | Not measured | Only GPT-5.6 Luna measured |
| Tax Agent Bench (Vals AI) | 24.8% | Not measured | Only GPT-5.6 Luna measured |
| Legal Research Bench (Vals AI) | 36.5% | Not measured | Only GPT-5.6 Luna measured |
| Excel Modeling Benchmark (Vals AI) | 67.1% | Not measured | Only GPT-5.6 Luna measured |
| BioMysteryBench (Vals AI) | 61.5% | Not measured | Only GPT-5.6 Luna measured |
| Terminal-Bench-Science (Vals AI) | 0.0% | Not measured | Only GPT-5.6 Luna measured |
| Vals Science Index | 25.3% | Not measured | Only GPT-5.6 Luna measured |
| τ²-Bench Telecom | Not measured | 17.8% | Only Llama 4 Maverick measured |
| APEX-Agents | 43.0% | Not measured | Only GPT-5.6 Luna measured |
| BALROG | 45.6% | Not measured | Only GPT-5.6 Luna measured |
| Vending-Bench 2 | $4,095 | Not measured | Only GPT-5.6 Luna measured |
| Neither model has a result on the other 23 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index120
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | 69.4% | GPT-5.6 Luna |
| LiveBench reasoning | 85.6% | Not measured | Only GPT-5.6 Luna measured |
| MysteryMechanism (Vals AI) | 14.4% | Not measured | Only GPT-5.6 Luna measured |
| MMMU Pro (Vals AI) | 85.0% | 71.7% | GPT-5.6 Luna |
| GPQA Diamond (Artificial Analysis run) | 91.1% | 67.1% | GPT-5.6 Luna |
| Humanity's Last Exam (Artificial Analysis run) | 39.5% | 4.9% | GPT-5.6 Luna |
| GPQA Diamond (Epoch AI run) | 91.6% | Not measured | Only GPT-5.6 Luna measured |
| ARC-AGI | 88.0% | 4.4% | GPT-5.6 Luna |
| ARC-AGI-2 | 59.5% | 0.0% | GPT-5.6 Luna |
| SimpleBench | 46.8% | 27.7% | GPT-5.6 Luna |
| Chess puzzles | 43.0% | Not measured | Only GPT-5.6 Luna measured |
| Mystery game puzzles | 21.0% | Not measured | Only GPT-5.6 Luna measured |
| LMArena Text: hard prompts | 1451 | 1281 | GPT-5.6 Luna |
| LMArena Text: expert | 1478 | 1259 | GPT-5.6 Luna |
| LMArena Text: hard prompts in English | 1460 | 1295 | GPT-5.6 Luna |
| LMArena Vision: diagrams | 1273 | 1134 | GPT-5.6 Luna |
| LMArena Vision: homework | 1282 | 1164 | GPT-5.6 Luna |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| AIME | Not measured | 25.2% | Only Llama 4 Maverick measured |
| MATH 500 (Vals AI) | Not measured | 85.2% | Only Llama 4 Maverick measured |
| MGSM (Vals AI) | Not measured | 92.4% | Only Llama 4 Maverick measured |
| AA Math Index | Not measured | 19.3 | Only Llama 4 Maverick measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 19.3% | Only Llama 4 Maverick measured |
| OTIS Mock AIME | 98.3% | Not measured | Only GPT-5.6 Luna measured |
| FrontierMath, tiers 1 to 3 | 82.1% | Not measured | Only GPT-5.6 Luna measured |
| FrontierMath, tier 4 | 61.0% | Not measured | Only GPT-5.6 Luna measured |
| ProofBench | 60.0% | Not measured | Only GPT-5.6 Luna measured |
| LMArena Text: math | 1459 | 1299 | GPT-5.6 Luna |
| LMArena Text: mathematical work | 1470 | 1295 | GPT-5.6 Luna |
| Neither model has a result on the other 2 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index100
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| MMLU Pro | 86.0% | 79.4% | GPT-5.6 Luna |
| LegalBench (Vals AI) | 84.0% | 77.8% | GPT-5.6 Luna |
| CorpFin (Vals AI) | 64.2% | 49.7% | GPT-5.6 Luna |
| TaxEval (Vals AI) | 76.2% | 66.6% | GPT-5.6 Luna |
| MortgageTax (Vals AI) | 67.3% | 58.5% | GPT-5.6 Luna |
| MedQA (Vals AI) | Not measured | 43.3% | Only Llama 4 Maverick measured |
| MedCode (Vals AI) | 42.4% | 36.5% | GPT-5.6 Luna |
| MMLU Pro (Artificial Analysis run) | Not measured | 80.9% | Only Llama 4 Maverick measured |
| SimpleQA Verified | 41.0% | Not measured | Only GPT-5.6 Luna measured |
| GeoBench | Not measured | 52.0% | Only Llama 4 Maverick measured |
| LMArena Text: science | 1433 | 1288 | GPT-5.6 Luna |
| LMArena Text: medicine and healthcare | 1399 | 1275 | GPT-5.6 Luna |
| LMArena Text: legal and government | 1446 | 1286 | GPT-5.6 Luna |
| LMArena Text: business and finance | 1439 | 1260 | GPT-5.6 Luna |
| Neither model has a result on the other 3 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| Design Arena: websites | 1256 | 871 | GPT-5.6 Luna |
| Design Arena: UI components | 1250 | 899 | GPT-5.6 Luna |
| Design Arena: game dev | 1196 | 838 | GPT-5.6 Luna |
| Design Arena: data viz | 1212 | 882 | GPT-5.6 Luna |
| Design Arena: 3D | 1121 | 901 | GPT-5.6 Luna |
| Design Arena: all code categories | 1231 | 880 | GPT-5.6 Luna |
| Design Arena: SVG | 1193 | Not measured | Only GPT-5.6 Luna measured |
| Design Arena: ASCII art | 1144 | Not measured | Only GPT-5.6 Luna measured |
| LiveBench design | 76.0% | Not measured | Only GPT-5.6 Luna measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index280
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | Leader |
|---|---|---|---|
| LiveBench data analysis | 78.0% | Not measured | Only GPT-5.6 Luna measured |
| Public Benefits Bench (Vals AI) | 61.2% | Not measured | Only GPT-5.6 Luna measured |
| SAGE (Vals AI) | 44.2% | 35.7% | GPT-5.6 Luna |
| MedScribe (Vals AI) | 84.4% | 54.2% | GPT-5.6 Luna |
| Vals Multimodal Index | 69.1% | Not measured | Only GPT-5.6 Luna measured |
| IFBench | Not measured | 43.0% | Only Llama 4 Maverick measured |
| Long context reasoning (AA-LCR) | 83.7% | 50.0% | GPT-5.6 Luna |
| DTBench | 89.1% | 61.9% | GPT-5.6 Luna |
| LMCA | 48.5% | 15.9% | GPT-5.6 Luna |
| Fiction.LiveBench | Not measured | 46.2% | Only Llama 4 Maverick measured |
| Creative writing (Lech Mazur) | Not measured | 62.0% | Only Llama 4 Maverick measured |
| Surface Evolver Bench | 61.9% | Not measured | Only GPT-5.6 Luna measured |
| Furniture assembly | 42.5% | Not measured | Only GPT-5.6 Luna measured |
| LMArena Text | 1431 | 1287 | GPT-5.6 Luna |
| LMArena Text: instruction following | 1437 | 1267 | GPT-5.6 Luna |
| LMArena Text: creative writing | 1396 | 1266 | GPT-5.6 Luna |
| LMArena Vision | 1257 | 1142 | GPT-5.6 Luna |
| LMArena Text: English | 1441 | 1303 | GPT-5.6 Luna |
| LMArena Text: non-English | 1417 | 1269 | GPT-5.6 Luna |
| LMArena Text: Chinese | 1470 | 1277 | GPT-5.6 Luna |
| LMArena Text: German | 1455 | 1291 | GPT-5.6 Luna |
| LMArena Text: Japanese | 1415 | 1206 | GPT-5.6 Luna |
| LMArena Text: Korean | 1415 | 1203 | GPT-5.6 Luna |
| LMArena Text: French | 1458 | 1259 | GPT-5.6 Luna |
| LMArena Text: Spanish | 1446 | 1293 | GPT-5.6 Luna |
| LMArena Text: Russian | 1427 | 1286 | GPT-5.6 Luna |
| LMArena Text: Polish | 1405 | 1275 | GPT-5.6 Luna |
| LMArena Text: multi-turn | 1434 | 1289 | GPT-5.6 Luna |
| LMArena Text: longer queries | 1436 | 1280 | GPT-5.6 Luna |
| LMArena Text: writing and language | 1419 | 1277 | GPT-5.6 Luna |
| LMArena Text: entertainment and media | 1402 | 1250 | GPT-5.6 Luna |
| LMArena Vision: English | 1263 | 1150 | GPT-5.6 Luna |
| LMArena Vision: Chinese | 1294 | 1141 | GPT-5.6 Luna |
| LMArena Vision: reading text | 1264 | 1143 | GPT-5.6 Luna |
| LMArena Vision: humor | 1190 | 1100 | GPT-5.6 Luna |
| LMArena Vision: creative writing | 1257 | 1099 | GPT-5.6 Luna |
| LMArena Document | 1457 | Not measured | Only GPT-5.6 Luna measured |
| Neither model has a result on the other 9 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.6 Luna | Llama 4 Maverick | 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
Llama 4 Maverick costs less
Repository review
50K input + 3K output tokens
Llama 4 Maverick costs less
Agent loop
200K input + 10K output tokens
Llama 4 Maverick 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 | OpenAI GPT-5.6 Luna | Meta Llama 4 Maverick |
|---|---|---|
| Maker | OpenAI | Meta Llama |
| Weights | No public weights listed | Open weights |
| Released | July 9, 2026 | April 5, 2025 |
| Context window | 1.05M | 524K |
| Input $/Mtok | $0.20 | $0.1875 |
| Output $/Mtok | $1.20 | $0.6525 |
| Coding index | 61.5 | 18.1 |
| Independent measurements behind it | 2 | 2 |
| Cost per solved task | $0.04 | $1.491 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.6 Luna better than Llama 4 Maverick for coding?
Pick GPT-5.6 Luna for coding: it scores 61.5 on the Vibe Coding Index against 18.1 for Llama 4 Maverick. Pick Llama 4 Maverick when output tokens cost 1.8x less.
Which is cheaper, GPT-5.6 Luna or Llama 4 Maverick?
GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens; Llama 4 Maverick costs $0.1875 and $0.6525. Per solved SWE-bench Verified task, GPT-5.6 Luna costs $0.04 and Llama 4 Maverick $1.491.
Which has the bigger context window, GPT-5.6 Luna or Llama 4 Maverick?
GPT-5.6 Luna, with 1.05M tokens against 524K.
How many benchmarks is the GPT-5.6 Luna vs Llama 4 Maverick comparison based on?
67 benchmarks have a result for both models as of October 21, 2026. GPT-5.6 Luna's coding index rests on 2 independent benchmarks and Llama 4 Maverick's on 2. Where only one model has a result, the page says so instead of naming a winner.