Doubao Seed Code — benchmark results
ByteDance's agentic-coding Doubao Seed model with a 256K context. Provider: ByteDance. Released 2025-12-01. Access: API.
Unified ELO 1690 ± 21, rank #291 of 1776 rated models, from 57 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA MMLU-Pro | 85.39 | Accuracy (%) | 92.4 |
| AA LiveCodeBench | 76.61 | Pass@1 (%) | 87.7 |
| AA-LCR | 65.3 | Score (self-reported) | 84.8 |
| AA Long Context Reasoning | 65.33 | Accuracy (%) | 82.5 |
| AA Omniscience - Software Engineering (SWE) - Julia | 32 | Accuracy (%) | 80.6 |
| AA SciCode | 40.74 | Accuracy (%) | 76.4 |
| AA Omniscience - Software Engineering (SWE) - Java | 27 | Accuracy (%) | 76.1 |
| AA AIME 2025 | 79.33 | Accuracy (%) | 74.4 |
| AA Omniscience - Science, Engineering & Mathematics | 34.4 | Accuracy (%) | 74.4 |
| AA Omniscience - Software Engineering (SWE) - Dart | 30 | Accuracy (%) | 73.5 |
| AA Omniscience - Business | 22.8 | Accuracy (%) | 73.3 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 40.91 | Accuracy (%) | 72 |
Interactive version: theaggregate.ai/model?slug=doubao-seed-code · How the rankings work · Data refreshed daily, snapshot 2026-07-22.