Model comparison
GPT-5.1 vs Llama 4 Maverick
GPT-5.1 is the stronger model overall, scoring 48.8 to 30.6 on the Noometry Index. Llama 4 Maverick costs 11× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Last verified . 41 shared benchmarks.
MetaSummary
- They share 41 benchmarks with published results for both. GPT-5.1 scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.1 leads 39.9 to 10.1.
- The biggest single-benchmark swing is ARC-AGI-1: 72.8% for GPT-5.1 and 4.4% for Llama 4 Maverick.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
- GPT-5.1 accepts more context: 400K tokens versus 128K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| GPT-5.1 | Llama 4 Maverick | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 48.8 | 30.6 |
| Released | 2025-11-13 | 2025-04-05 |
| Weights | Proprietary | Open |
| Context window | 400K | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $1.25 | $0.19 |
| Output $ / M tokens | $10 | $0.65 |
| Results tracked | 63 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
GPT-5.1: 46.9 (#59), Llama 4 Maverick: 26.3 (#318)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| SWE-bench Verified (bash only) | 66% | 21% |
| SciCode | 43.3% | 33.1% |
| WeirdML | 60.8% | 24.5% |
| LMArena Coding | 1454 | 1302 |
| ALE-Bench | 1,192 | 172.97 |
| SWE-bench Verified | 68% | — |
| Aider Polyglot | — | 15.6% |
| LMArena WebDev | 1395 | — |
| GSO | 13.7% | — |
| BigCodeBench Instruct | — | 49.7% |
| LiveBench Coding | 72.5% | — |
| BigCodeBench Complete | — | 61.4% |
Agentic & Tool Use GPT-5.1 leads
GPT-5.1: 32.5 (#55), Llama 4 Maverick: 27.9 (#86)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| Terminal-Bench | 47.6% | — |
| Berkeley Function Calling Leaderboard | — | 37.3% |
| DeepResearch Bench | 42.8% | — |
| LMArena Search | 1199 | — |
| Vending-Bench 2 | 1,473 | — |
Reasoning GPT-5.1 leads
GPT-5.1: 39.9 (#50), Llama 4 Maverick: 10.1 (#337)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| ARC-AGI-2 | 17.6% | 0% |
| SimpleBench | 53.2% | 27.7% |
| ARC-AGI-1 | 72.8% | 4.4% |
| CritPt | 4.9% | 0% |
| EnigmaEval | 11.2% | 0.6% |
| LMArena Hard Prompts | 1457 | 1281 |
| DTBench | 90.1% | 61.9% |
| LMCA | 43.9% | 15.9% |
| Epoch Capabilities Index | 149.66 | 132.2 |
| ForecastBench | 58.1 | 57.5 |
| Kagi LLM Benchmark | — | 55.9% |
| NYT Connections (extended) | — | 8% |
| Chess Puzzles | 32% | — |
| LiveBench Reasoning | 95.8% | — |
| Mystery Game Puzzles | 19% | — |
| LiveBench Data Analysis | 72.1% | — |
| LiveBench | 78.8% | — |
Math GPT-5.1 leads
GPT-5.1: 50.3 (#45), Llama 4 Maverick: 24.6 (#256)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 20.6% |
| Omni-MATH | 46.4% | 42.2% |
| LMArena Math | 1446 | 1299 |
| FrontierMath (Feb 2025 set) | 31% | 0.7% |
| LiveBench Math | 94.5% | — |
| MATH Level 5 | — | 73% |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GPT-5.1 leads
GPT-5.1: 50.5 (#65), Llama 4 Maverick: 33.3 (#196)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 87.6% | 67% |
| Humanity's Last Exam | 23.7% | 5.7% |
| MMLU-Pro | 57.9% | 81% |
| Vectara Hallucination Rate | 10.9% | 8.2% |
| GPQA (HELM) | 44.2% | 65% |
| LMArena Expert | 1470 | 1259 |
| SimpleQA Verified | 48% | — |
| Confabulations | — | 22.6% |
Multimodal GPT-5.1 leads
GPT-5.1: 44.0 (#17), Llama 4 Maverick: 30.9 (#100)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1250 | 1141 |
| GeoBench | — | 52% |
| VPCT | 58.7% | — |
| LMArena Document | 1403 | — |
| SpatialViz-Bench | — | 31.8% |
Multilingual GPT-5.1 leads
GPT-5.1: 54.0 (#51), Llama 4 Maverick: 42.4 (#190)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1431 | 1269 |
| LMArena Chinese | 1495 | 1277 |
| LMArena French | 1451 | 1258 |
| LMArena German | 1439 | 1291 |
| LMArena Japanese | 1453 | 1207 |
| LMArena Korean | 1401 | 1203 |
| LMArena Russian | 1435 | 1286 |
| LMArena Spanish | 1430 | 1293 |
Instruction Following GPT-5.1 leads
GPT-5.1: 84.0 (#1), Llama 4 Maverick: 71.9 (#140)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| IFEval | 93.5% | 90.8% |
| LMArena Instruction Following | 1442 | 1267 |
| LiveBench Instruction Following | 93.3% | — |
Long Context GPT-5.1 leads
GPT-5.1: 47.6 (#14), Llama 4 Maverick: 31.4 (#274)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1448 | 1280 |
| Fiction.LiveBench | — | 46.2% |
| CL-bench | 23.7% | — |
| CL-bench Life | 17.3% | — |
Writing & Preference GPT-5.1 leads
GPT-5.1: 64.7 (#53), Llama 4 Maverick: 39.0 (#247)
| Benchmark | GPT-5.1 | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1443 | 1287 |
| LMArena Creative Writing | 1427 | 1266 |
| WildBench | 86.3% | 80% |
| LMArena Multi-Turn | 1449 | 1289 |
| Short-Story Creative Writing | — | 62% |
| EQ-Bench Creative Writing | — | 860 |
| LiveBench Language | 80.2% | — |
Frequently asked questions
Is GPT-5.1 better than Llama 4 Maverick?
GPT-5.1 is the stronger model overall, scoring 48.8 to 30.6 on the Noometry Index. Llama 4 Maverick costs 11× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Which is cheaper, GPT-5.1 or Llama 4 Maverick?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; GPT-5.1 lists at $1.25 and $10.
Is GPT-5.1 or Llama 4 Maverick better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.9 versus 26.3 in the Noometry coding category.
Which has the bigger context window?
GPT-5.1 does, with 400K tokens against 128K.
How many benchmarks do GPT-5.1 and Llama 4 Maverick share?
41 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and Llama 4 Maverick has 54.