Doubao-1.5-Thinking-Pro: benchmark results
Provider: ByteDance. Released 2025-04-17. Access: API.
Aggregate Rating 1719 ± 21, rank #260 of 10516 rated models, from 43 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLMEval-Fair - Economics | 9.67 | Discipline Score (0-10) | 100 |
| LLMEval-Fair - Education | 9.43 | Discipline Score (0-10) | 100 |
| LLMEval-Fair - History | 9.7 | Discipline Score (0-10) | 100 |
| LLMEval-Fair - Law | 9.77 | Discipline Score (0-10) | 100 |
| LLMEval-Fair - Literature | 8.93 | Discipline Score (0-10) | 100 |
| LLMEval-Fair - Medicine | 8.97 | Discipline Score (0-10) | 100 |
| LLMEval-Fair - Military | 8.97 | Discipline Score (0-10) | 100 |
| LLMEval-Fair - Overall | 93.67 | Absolute Score (0-100) | 100 |
| LLMEval-Fair - Science | 9.23 | Discipline Score (0-10) | 100 |
| MedErrBench (Chinese) - Error Detection | 75 | Detection accuracy (%): whether the note contains an error ( | 100 |
| MedErrBench (Chinese) - Error Localization | 73.5 | Localization accuracy (%): naming the sentence that contains | 100 |
| MedErrBench (English) - Error Correction | 76.6 | Correction ROUGE-1 (%): unigram overlap of the corrected sen | 100 |
Interactive version: theaggregate.ai/model?slug=doubao-1-5-thinking-pro · How It Works · Data refreshed daily, snapshot 2026-11-21.