Model comparison
Claude 3 Opus vs Llama 4 Maverick
Llama 4 Maverick is the stronger model overall, scoring 30.6 to 29.3 on the Noometry Index.
Last verified . 31 shared benchmarks.
MetaSummary
- They share 31 benchmarks with published results for both. Claude 3 Opus scores higher in 4 categories and Llama 4 Maverick in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Llama 4 Maverick leads 24.6 to 13.9.
- The biggest single-benchmark swing is MATH Level 5: 37.5% for Claude 3 Opus and 73% for Llama 4 Maverick.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | Llama 4 Maverick | |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 29.3 | 30.6 |
| Released | 2024-02-29 | 2025-04-05 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.19 |
| Output $ / M tokens | — | $0.65 |
| Results tracked | 46 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3 Opus leads
Claude 3 Opus: 32.4 (#261), Llama 4 Maverick: 26.3 (#318)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| WeirdML | 19.2% | 24.5% |
| BigCodeBench Instruct | 45.5% | 49.7% |
| LMArena Coding | 1264 | 1302 |
| BigCodeBench Complete | 57.4% | 61.4% |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| SciCode | — | 33.1% |
| LiveBench Coding | 38.6% | — |
| ALE-Bench | — | 172.97 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Llama 4 Maverick leads
Claude 3 Opus: 24.3 (#110), Llama 4 Maverick: 27.9 (#86)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.3% |
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
Reasoning Claude 3 Opus leads
Claude 3 Opus: 14.2 (#318), Llama 4 Maverick: 10.1 (#337)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| SimpleBench | 23.5% | 27.7% |
| EnigmaEval | 0.8% | 0.6% |
| LMArena Hard Prompts | 1246 | 1281 |
| DTBench | 61.6% | 61.9% |
| LMCA | 17% | 15.9% |
| Epoch Capabilities Index | 126.91 | 132.2 |
| ForecastBench | 58.4 | 57.5 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 55.9% |
| NYT Connections (extended) | — | 8% |
| ARC-AGI-1 | — | 4.4% |
| CritPt | — | 0% |
| Chess Puzzles | 5% | — |
| LiveBench Reasoning | 40.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math Llama 4 Maverick leads
Claude 3 Opus: 13.9 (#293), Llama 4 Maverick: 24.6 (#256)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 20.6% |
| LMArena Math | 1273 | 1299 |
| MATH Level 5 | 37.5% | 73% |
| Omni-MATH | — | 42.2% |
| LiveBench Math | 43.6% | — |
| FrontierMath (Feb 2025 set) | — | 0.7% |
Knowledge Llama 4 Maverick leads
Claude 3 Opus: 24.5 (#262), Llama 4 Maverick: 33.3 (#196)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 47.2% | 67% |
| Confabulations | 22.7% | 22.6% |
| LMArena Expert | 1223 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| SimpleQA Verified | 12.6% | — |
| MMLU-Pro | — | 81% |
| Vectara Hallucination Rate | — | 8.2% |
| GPQA (HELM) | — | 65% |
| MMLU | 84.6% | — |
Multimodal Llama 4 Maverick leads
Claude 3 Opus: 26.6 (#112), Llama 4 Maverick: 30.9 (#100)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1023 | 1141 |
| GeoBench | — | 52% |
| SpatialViz-Bench | — | 31.8% |
Multilingual Too close to call
Claude 3 Opus: 41.6 (#202), Llama 4 Maverick: 42.4 (#190)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1258 | 1269 |
| LMArena Chinese | 1248 | 1277 |
| LMArena French | 1274 | 1258 |
| LMArena German | 1258 | 1291 |
| LMArena Japanese | 1204 | 1207 |
| LMArena Korean | 1187 | 1203 |
| LMArena Russian | 1280 | 1286 |
| LMArena Spanish | 1246 | 1293 |
Instruction Following Llama 4 Maverick leads
Claude 3 Opus: 64.3 (#223), Llama 4 Maverick: 71.9 (#140)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1248 | 1267 |
| LiveBench Instruction Following | 63.9% | — |
| IFEval | — | 90.8% |
Long Context Claude 3 Opus leads
Claude 3 Opus: 38.2 (#197), Llama 4 Maverick: 31.4 (#274)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1259 | 1280 |
| Fiction.LiveBench | — | 46.2% |
Writing & Preference Claude 3 Opus leads
Claude 3 Opus: 47.4 (#208), Llama 4 Maverick: 39.0 (#247)
| Benchmark | Claude 3 Opus | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1262 | 1287 |
| LMArena Creative Writing | 1235 | 1266 |
| LMArena Multi-Turn | 1275 | 1289 |
| Short-Story Creative Writing | — | 62% |
| EQ-Bench Creative Writing | — | 860 |
| WildBench | — | 80% |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than Llama 4 Maverick?
Llama 4 Maverick is the stronger model overall, scoring 30.6 to 29.3 on the Noometry Index.
Is Claude 3 Opus or Llama 4 Maverick better for coding?
Claude 3 Opus scores higher on coding benchmarks: 32.4 versus 26.3 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and Llama 4 Maverick share?
31 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Llama 4 Maverick has 54.