Sailor2-8B
SEA-first multilingual LLM built on Qwen2.5, Apache 2.0.
About
Sailor2-8B is the 8B member of the Sailor2 family from Sea AI Lab, described in the technical report published February 18, 2025 (arXiv 2502.12982). It continues pre-training Qwen2.5 on 500B tokens, 400B of them Southeast-Asia-specific, covering 13 to 15 SEA languages including Thai, Vietnamese, Indonesian, Tagalog, Burmese, Khmer, and Lao.
The family is released under Apache 2.0 in 1B, 8B, and 20B sizes. The report claims the 20B variant reaches a 50-50 win rate against GPT-4o across SEA languages, and the 8B hits the sweet spot for production deployment of regional chat and content workloads.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 125; Trending score 78/100 from researcher assessment.