Seed-X-PPO-7B — benchmark results
Provider: ByteDance. Released 2025-07-18. Access: Open.
Unified ELO 1421 ± 36, rank #1136 of 1839 rated models, from 24 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DITING - Idiom Translation | 5.42 | Score | 100 |
| DITING | 4.65 | Score | 84.6 |
| DITING - Lexical Ambiguity | 5.46 | Score | 84.6 |
| DITING - Terminology Localization | 4.52 | Score | 76.9 |
| LatamBoard - Spanish OpenBookQA | 35.8 | Score (%) | 71.2 |
| LatamBoard - Spanish COPA | 82.2 | Score (%) | 68.2 |
| LatamBoard - Spanish TeleIA | 66.67 | Score (%) | 62.1 |
| DITING - Tense Consistency | 4.64 | Score | 46.2 |
| DITING - Zero Pronoun Translation | 3.68 | Score | 46.2 |
| LatamBoard - Spanish Score | 57.1 | Score (%) | 39.4 |
| LatamBoard - Spanish MGSM | 4.4 | Score (%) | 36.4 |
| LatamBoard - Spanish PAWS | 60 | Score (%) | 36.4 |
Interactive version: theaggregate.ai/model?slug=seed-x-ppo-7b · How It Works · Data refreshed daily, snapshot 2026-08-05.