Doubao-1.5-Pro — benchmark results
ByteDance's Doubao 1.5 Pro, a closed general-purpose model from the Doubao family. Provider: ByteDance. Released 2025-01-15. Access: API.
Unified ELO 1659 ± 14, rank #347 of 1776 rated models, from 125 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LMM Reasoning - DynaMath - Level-high school; Avg | 73.2 | Accuracy (%) | 100 |
| Open LMM Reasoning - DynaMath - Subject-analytic geometry; Avg | 67.7 | Accuracy (%) | 100 |
| Open LMM Reasoning - DynaMath - Subject-plane geometry; Avg | 67.4 | Accuracy (%) | 100 |
| Open LMM Reasoning - DynaMath - Subject-scientific figure; Avg | 75.6 | Accuracy (%) | 100 |
| Open LMM Reasoning - LogicVista - deductive | 88.2 | Accuracy (%) | 100 |
| Open LMM Reasoning - LogicVista - numerical | 91.6 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVerse - Angle | 68.4 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVerse - Length | 68.7 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVerse - Plane Geometry | 67.5 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVision | 51.5 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVision - logic | 52.1 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVision - metric geometry - angle | 67.6 | Accuracy (%) | 100 |
Interactive version: theaggregate.ai/model?slug=doubao-1-5-pro · How the rankings work · Data refreshed daily, snapshot 2026-07-22.