Llama-SEA-LION-v3-8B-IT: benchmark results

Provider: AI Singapore. Released 2024-12-18. Access: Open.

Aggregate Rating 1470 ± 27, rank #2946 of 10524 rated models, from 144 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
SEA-HELM (Vietnamese) - Sentiment Analysis63.88Normalized Score93.2
SEA-HELM (Indonesian) - SEA-Safeguard73.79Normalized Score87.8
SEA-HELM (Malay) - Toxicity Detection14.36Normalized Score87.8
SEA-HELM (Thai) - Summarization23.93Normalized Score82.4
SEA-HELM (Indonesian) - Summarization17.15Normalized Score78.4
SEA-HELM (Malay) - Safety41.15Normalized Score68.9
ThaiSafetyBench - Malicious Uses10.29Attack Success Rate (%)65.2
SEA-HELM (Tamil) - Summarization11.46Normalized Score63.5
SEA-HELM (Vietnamese) - Summarization16.3Normalized Score62.2
ThaiSafetyBench - Discrimination, Exclusion, Toxicity, Hateful, Offensive8.57Attack Success Rate (%)60.9
SEA-HELM (Filipino) - SEA-Safeguard67.87Normalized Score58.1
ThaiSafetyBench - Misinformation Harms20.91Attack Success Rate (%)56.5

Interactive version: theaggregate.ai/model?slug=llama-sea-lion-v3-8b-it · How It Works · Data refreshed daily, snapshot 2026-11-22.