Google: lyria-2
1 benchmarks
Leaderboard position
#741
Overall rank (of 1121 models)
594
Elo rating
1
Benchmarks
Top 66.1%
Percentile
In Google (of 72 models)
Above: Qwen: qwen3-vl-4b-instruct #740
Below: JetBrains: mellum2-12b-a2.5b-base #742
Benchmark highlights
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 | ~81.87 | 50% | — |
| AIME 2025 | ~57.56 | 50% | — |
| AIME 2026 | ~68.46 | 50% | — |
| HMMT Feb 2026 | ~59.95 | 50% | — |
| GPQA Diamond | ~52.35 | 50% | — |
| MMLU-Pro | ~54.83 | 50% | — |
| Humanity's Last Exam (no tools) | ~7 | 50% | — |
| IFEval | ~78.72 | 50% | — |
| IFBench | ~49.03 | 50% | — |
| HumanEval | ~66.15 | 50% | — |
| MBPP | ~67.13 | 50% | — |
| Terminal-Bench v2.1 | ~12.48 | 50% | — |
| MMLU | ~60.96 | 50% | — |
| SWE-bench Verified | ~31.02 | 50% | — |
| SWE-bench Pro | ~10.52 | 50% | — |
| SciCode | ~22.85 | 50% | — |
| AA-LCR | ~38.01 | 50% | — |
| Artificial Analysis Intelligence Index v5.0.1 | ~12 | 50% | — |
| Humanity's Last Exam (with tools) | ~11.9 | 50% | — |
| GSM8K | ~73.03 | 50% | — |
| -(CoderPPL) | ~-2.54 | 50% | — |
| LiveCodeBench v6 | ~49.98 | 50% | — |
| MMMU-Pro | ~48.06 | 50% | — |
| MATH | ~49.83 | 50% | — |
| BBH | ~59.55 | 50% | — |