Model comparison
GPT-6 Luna vs Llama 4 Maverick
GPT-6 Luna is the stronger model overall, scoring 53.3 to 30.9 on the Noometry Index.
Last verified . 29 shared benchmarks.
MetaSummary
- They share 29 benchmarks with published results for both. GPT-6 Luna scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 26.0.
- The biggest single-benchmark swing is ARC-AGI-1: 86.7% for GPT-6 Luna and 4.4% for Llama 4 Maverick.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
- GPT-6 Luna accepts more context: 1.05M tokens versus 128K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Luna | Llama 4 Maverick | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 53.3 | 30.9 |
| Released | 2026-09-22 | 2025-04-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $0.10 | $0.19 |
| Output $ / M tokens | $0.50 | $0.65 |
| Results tracked | 42 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 55.5 (#25), Llama 4 Maverick: 26.6 (#324)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| SciCode | 54.6% | 33.1% |
| LMArena Coding | 1439 | 1302 |
| ALE-Bench | 1,577 | 172.97 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| LMArena WebDev | 1581 | — |
| WeirdML | — | 24.5% |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |
Agentic & Tool Use GPT-6 Luna leads
GPT-6 Luna: 33.3 (#54), Llama 4 Maverick: 28.2 (#91)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| APEX-Agents | 44.3% | — |
| Berkeley Function Calling Leaderboard | — | 37.3% |
| GDP.pdf | 23% | — |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 48.2 (#41), Llama 4 Maverick: 10.1 (#342)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| ARC-AGI-2 | 59.3% | 0% |
| NYT Connections (extended) | 68.7% | 8% |
| ARC-AGI-1 | 86.7% | 4.4% |
| CritPt | 19.4% | 0% |
| LMArena Hard Prompts | 1411 | 1281 |
| DTBench | 90.1% | 61.9% |
| LMCA | 44.5% | 15.9% |
| Epoch Capabilities Index | 156.28 | 132.2 |
| SimpleBench | — | 27.7% |
| Kagi LLM Benchmark | — | 55.9% |
| Chess Puzzles | 31% | — |
| EnigmaEval | — | 0.6% |
| Mystery Game Puzzles | 7% | — |
| ForecastBench | — | 57.5 |
Math GPT-6 Luna leads
GPT-6 Luna: 76.1 (#15), Llama 4 Maverick: 26.0 (#262)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 20.6% |
| LMArena Math | 1416 | 1299 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| ProofBench | 64% | — |
| Omni-MATH | — | 42.2% |
| MATH Level 5 | — | 73% |
| FrontierMath (Feb 2025 set) | — | 0.7% |
Knowledge GPT-6 Luna leads
GPT-6 Luna: 57.0 (#41), Llama 4 Maverick: 33.4 (#204)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 90.5% | 67% |
| LMArena Expert | 1444 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| SimpleQA Verified | 41.4% | — |
| MMLU-Pro | — | 81% |
| Confabulations | — | 22.6% |
| Vectara Hallucination Rate | — | 8.2% |
| GPQA (HELM) | — | 65% |
Multimodal GPT-6 Luna leads
GPT-6 Luna: 42.4 (#30), Llama 4 Maverick: 31.6 (#105)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1217 | 1142 |
| GeoBench | — | 52% |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
| SpatialViz-Bench | — | 31.8% |
Multilingual GPT-6 Luna leads
GPT-6 Luna: 50.5 (#117), Llama 4 Maverick: 42.2 (#195)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1386 | 1269 |
| LMArena Chinese | 1433 | 1277 |
| LMArena French | 1420 | 1259 |
| LMArena German | 1369 | 1291 |
| LMArena Japanese | 1369 | 1207 |
| LMArena Korean | 1360 | 1203 |
| LMArena Russian | 1394 | 1286 |
| LMArena Spanish | 1393 | 1293 |
Instruction Following GPT-6 Luna leads
GPT-6 Luna: 74.3 (#99), Llama 4 Maverick: 71.7 (#146)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1409 | 1267 |
| IFEval | — | 90.8% |
Long Context GPT-6 Luna leads
GPT-6 Luna: 43.0 (#111), Llama 4 Maverick: 31.4 (#279)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1409 | 1280 |
| Fiction.LiveBench | — | 46.2% |
Writing & Preference GPT-6 Luna leads
GPT-6 Luna: 58.3 (#119), Llama 4 Maverick: 38.8 (#252)
| Benchmark | GPT-6 Luna | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1391 | 1287 |
| LMArena Creative Writing | 1363 | 1267 |
| LMArena Multi-Turn | 1396 | 1289 |
| Short-Story Creative Writing | — | 62% |
| EQ-Bench Creative Writing | — | 860 |
| WildBench | — | 80% |
Frequently asked questions
Is GPT-6 Luna better than Llama 4 Maverick?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 30.9 on the Noometry Index.
Which is cheaper, GPT-6 Luna or Llama 4 Maverick?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.
Is GPT-6 Luna or Llama 4 Maverick better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 128K.
How many benchmarks do GPT-6 Luna and Llama 4 Maverick share?
29 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Llama 4 Maverick has 54.