Doubao-Seed-1.6 — benchmark results

Provider: ByteDance. Released 2025-06-11. Access: Open.

Unified ELO 1564 ± 25, rank #648 of 1839 rated models, from 27 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LiveSecBench70.83Overall Score (%)83.3
Emotional intelligence in large language model61.2Total Kappa (self-reported)66.7
VitaBench 2.037.3Avg@4 Full Context (self-reported)52.6
QZhou Flowchart QA83.83Accuracy (%)50
VerdictBench56.26Accuracy (self-reported)50
LLM2014 Vision 2025-0735.32Median Score46.7
LongWebBench4.77Overall (self-reported)33.3
LLM2014 Code 2025-0734.99Median Score23.5
StressEval28.1Overall (self-reported)22.2
LLM2014 Code 2025-09 - TypeScript5.63Score21.1
LLM2014 Code 2025-11 - Golang3Score20
LLM2014 Code 2025-11 - C#4.23Score16

Interactive version: theaggregate.ai/model?slug=doubao-seed-1-6 · How It Works · Data refreshed daily, snapshot 2026-08-05.