Data Prediction on DSPredict Hard (Private test)
85.7Valid ScoreGPT-5.1
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-5.1reasoning=high2026.01 | 85.7 | 4.8 | 14.3 | |
| Claude Sonnet 42026.01 | 85.7 | 2.4 | 4.8 | |
| GPT-5.1reasoning=medium2026.01 | 81 | 4.8 | 7.1 | |
| Deepseek v3.12026.01 | 76.2 | 2.4 | 7.1 | |
| Claude Sonnet 4.52026.01 | 71.4 | 0 | 4.8 | |
| GPT-5.1reasoning=none2026.01 | 69 | 2.4 | 10.3 | |
| Kimi K2 Instruct2026.01 | 69 | 0 | 0 | |
| Qwen3-Coderparameters=480B2026.01 | 66.7 | 2.4 | 5.9 | |
| Qwen3 Instructparameters=235B2026.01 | 64.3 | 2.4 | 2.4 | |
| GPT-5reasoning=medium2026.01 | 52.4 | 0 | 2.4 | |
| Qwen3 Instructparameters=4B2026.01 | 40.5 | 0 | 0 | |
| Qwen2.5 Instructparameters=7B2026.01 | 4.8 | 0 | 0 |