Search on HLE text
54.2ScoreSeed2.0 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed2.0 Pro2026.06 | 54.2 | |
| Gemini-3-pro High2026.06 | 46.9 | |
| Gemini-3-ProParams=-, Evaluation Tools=Same as ours, Source=[28]2026.02 | 45.8 | |
| GPT-5.2 High2026.06 | 45.5 | |
| Claude-Opus-4.52026.06 | 43.2 | |
| GPT-5Params=-2026.02 | 41.7 | |
| Seed1.8Params=-2026.02 | 40.9 | |
| Claude-Sonnet-4.52026.06 | 32 | |
| REDSearcher-MM-RLParams=30B2026.02 | 25.3 | |
| REDSearcher-MM-SFTParams=30B2026.02 | 24.4 | |
| Qwen3-VL ThinkingParams=235B2026.02 | 14.5 | |
| Qwen3-VL ThinkingParams=30B2026.02 | 8.8 |