- LiveCodeBench (10/01/2024-02/01/2025)
- MATH-500 95
- MMLU Pro 82.2
- GPQA Diamond 73.7
- Multilingual MMLU 85.8
- MMMU 76.1
Llama 4 Behemoth
Llama 4 Behemoth.
5
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Meta Apr 2025 Llama 4 Behemoth
Flagship
137.4
#202 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
- MMLU-Pro 82.2% #58 /167
- MMMLU 85.8% #28 /71
- MMMU 76.1% #18 /136
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 73.7% | #157 of 302 | Lab model card | |
| MATH-500 | 95% | #40 of 70 | Lab model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMMU | 76.1% | #18 of 136 | Official leaderboard |
Its lab's numbers put it at 137.4, number 202 of 404
The Self-Reported Index fits one score to the 5 capability results Meta reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 133.8 to 141.1. How we compute it.
- MATH-500 95
- MMMLU 85.8
- MMLU-Pro 82.2
- MMMU 76.1
- GPQA Diamond 73.7
Its lab's numbers next to 2 leaderboards
Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.