Google: veo-3-fast
1 benchmarks
Leaderboard position
#707
Overall rank (of 1121 models)
653
Elo rating
1
Benchmarks
Top 63.1%
Percentile
In Google (of 72 models)
Above: NVIDIA Corporation: parakeet-rnnt-1.1b #706
Below: Microsoft: phi-4-multimodal-instruct #708
Benchmark highlights
Best benchmark
Text-to-Video Arena (Arena.ai) (Sep 21, 2026)
1248
Worst benchmark
Text-to-Video Arena (Arena.ai) (Sep 21, 2026)
1248
Nearest competitor from another company
The model from another company with the most similar benchmark scores.
appeared as nearest in 1 of 1 benchmarks
1/1
100.0%
Benchmarks
Predicted scores
Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.
| Benchmark | Predicted | Confidence | Based on |
|---|---|---|---|
| MATH-500 | ~82.89 | 50% | — |
| AIME 2025 | ~59.49 | 50% | — |
| AIME 2026 | ~69.6 | 50% | — |
| HMMT Feb 2026 | ~61.09 | 50% | — |
| GPQA Diamond | ~53.65 | 50% | — |
| MMLU-Pro | ~56.28 | 50% | — |
| Humanity's Last Exam (no tools) | ~8.05 | 50% | — |
| IFEval | ~79.52 | 50% | — |
| IFBench | ~50 | 50% | — |
| HumanEval | ~67.86 | 50% | — |
| MBPP | ~69.38 | 50% | — |
| Terminal-Bench v2.1 | ~14.52 | 50% | — |
| MMLU | ~62.48 | 50% | — |
| SWE-bench Verified | ~32.87 | 50% | — |
| SWE-bench Pro | ~12.24 | 50% | — |
| SciCode | ~23.78 | 50% | — |
| AA-LCR | ~39.16 | 50% | — |
| Artificial Analysis Intelligence Index v5.0.1 | ~12 | 50% | — |
| Humanity's Last Exam (with tools) | ~11.9 | 50% | — |
| GSM8K | ~75.24 | 50% | — |
| -(CoderPPL) | ~-2.17 | 50% | — |
| LiveCodeBench v6 | ~51.33 | 50% | — |
| MMMU-Pro | ~49.17 | 50% | — |
| MATH | ~52.31 | 50% | — |
| BBH | ~60.86 | 50% | — |