Meta
Llama 4 Maverick is a Meta Llama 4 family model, from Meta's open-weight generation focused on multimodal intelligence and broad developer deployment. This page shows how the benchmarked variant ranks against hosted frontier models and other open-weight systems across quality, speed, and price context.
Introducing Llama 4Rank #343 across 526
Rank #256 across 436
Rank #134 across 357
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 14.3 | #343 |
| Artificial Analysis Coding Index | coding | 15.6 | #256 |
| Artificial Analysis Math Index | math | 19.3 | #211 |
| MMLU-Pro | reasoning | 80.9% | #96 |
| reasoning |
| 67.1% |
| #246 |
| Humanity's Last Exam | reasoning | 4.8% | #344 |
| LiveCodeBench | coding | 39.7% | #185 |
| SciCode | coding, reasoning | 33.1% | #236 |
| MATH-500 | math | 88.9% | #78 |
| AIME | math | 39.0% | #72 |
| Output Speed | speed | 92.9 tok/s | #147 |
| Time to First Token | speed | 0.59s | #64 |
| Blended Price | cost | $0.475/M | #134 |
| Input Price | cost | $0.350/M | #168 |
| Output Price | cost | $0.850/M | #122 |
| Value Index | cost, overall | 30.1 | #181 |