gwBenchmarks: leaderboard
Metric: Waveform (self-reported). Source: benchmarklist.com. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Haiku 4.5 | 59.31 |
| 2 | DeepSeek V4 Pro | 48.3 |
| 3 | Kimi K2.6 | 39.12 |
| 4 | GPT-5.4 Mini | 35.97 |
| 5 | Gemini 3.1 Pro (Preview) | 34.25 |
| 6 | Gemini 3 Flash | 21.27 |
| 7 | GPT-5.2 | 19.35 |
| 8 | GPT-5.5 (High) | 17.82 |
| 9 | GPT-5.3 Codex | 17.82 |
| 10 | Claude Opus 4.7 | 16.48 |
| 11 | Claude Opus 4.6 | 14.75 |
| 12 | Claude Sonnet 4.6 | 14.22 |
Interactive version: theaggregate.ai/benchmark?slug=gwbenchmarks · How It Works · Data refreshed daily, snapshot 2026-09-05.