perplexity: sonar-reasoning-pro
2 benchmarks
Leaderboard position
#448
Overall rank (of 1126 models)
1211
Elo rating
2
Benchmarks
Top 39.8%
Percentile
In perplexity (of 7 models)
Above: Amazon: nova-2.0-lite #447
Below: Qwen: qwen3-next-80b-a3b #449
Benchmark highlights
Nearest competitor from another company
The model from another company with the most similar benchmark scores.
appeared as nearest in 1 of 2 benchmarks
1/2
50.0%
1/2
50.0%
| Model | Votes | % of benchmarks |
|---|---|---|
| OpenAI: gpt-oss-20b | 1 | 50.0% |
| MiniMax: minimax-m1-80k | 1 | 50.0% |
Benchmarks
Predicted scores
Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.
| Benchmark | Predicted | Confidence | Based on |
|---|---|---|---|
| ARC-Challenge | ~81.25 | 50% | — |
| ARC-Easy | ~94.8 | 50% | — |
| HellaSwag | ~92.54 | 50% | — |
| PIQA | ~84.39 | 50% | — |
| WinoGrande | ~81.06 | 50% | — |
| Humanity's Last Exam (no tools) | ~17.9 | 50% | — |
| Artificial Analysis Intelligence Index v5.0.1 | ~12 | 50% | — |
| MATH-500 | ~93.52 | 50% | — |
| AIME 2025 | ~78.2 | 50% | — |
| AIME 2026 | ~80.22 | 50% | — |
| HMMT Feb 2026 | ~71.8 | 50% | — |
| GPQA Diamond | ~65.95 | 50% | — |
| MMLU-Pro | ~70.57 | 50% | — |
| IFEval | ~87.78 | 50% | — |
| IFBench | ~59.28 | 50% | — |
| HumanEval | ~95.47 | 50% | — |
| MBPP | ~93.7 | 50% | — |
| Terminal-Bench v2.1 | ~33.47 | 50% | — |
| MMLU | ~77.96 | 50% | — |
| SWE-bench Verified | ~50.55 | 50% | — |
| SWE-bench Pro | ~28.65 | 50% | — |
| SciCode | ~32.43 | 50% | — |
| AA-LCR | ~49.95 | 50% | — |
| Humanity's Last Exam (with tools) | ~24.4 | 50% | — |
| GSM8K | ~98.23 | 50% | — |