Sailor-0.5B-Chat: benchmark results
Provider: Sea AI Lab. Access: Open.
Unified ELO 1227 ± 19, rank #2852 of 2928 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 1.82 | Accuracy (%) (5-shot) | 18.5 |
| Open LLM Leaderboard v1 - MMLU | 26.73 | Accuracy (%) (5-shot) | 13.6 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 39.85 | MC2 (%) (0-shot) | 12.1 |
| Open LLM Leaderboard v1 - WinoGrande | 56.51 | Accuracy (%) (5-shot) | 9.9 |
| Open LLM Leaderboard v1 - ARC Challenge | 30.38 | Normalized accuracy (%) (25-shot) | 9.4 |
| Open LLM Leaderboard v1 - HellaSwag | 45.51 | Normalized accuracy (%) (10-shot) | 9.1 |
| SEA LLM Leaderboard - SeaExam - Indonesian (Public) | 27.5 | Public Accuracy (%) | 3.3 |
| SEA LLM Leaderboard - SeaExam - Thai (Public) | 26.9 | Public Accuracy (%) | 3.3 |
| SEA LLM Leaderboard - SeaExam (Public) | 26.7 | Public Average Score (%) | 2.2 |
| SEA LLM Leaderboard - SeaExam - Vietnamese (Private) | 28.5 | Private Accuracy (%) | 2.2 |
| SEA LLM Leaderboard - SeaExam | 26.3 | Private Average Score (%) | 1.7 |
| SEA LLM Leaderboard - SeaExam - Thai (Private) | 24.9 | Private Accuracy (%) | 1.7 |
Interactive version: theaggregate.ai/model?slug=sailor-0-5b-chat · How It Works · Data refreshed daily, snapshot 2026-09-23.