Named Models
SEA-LION-E5-Embedding-600MAI SingaporeReleasedOct 2026Runs onCPU serverClassembedding
AI Singapore SEA-LION-E5-Embedding-600M encoder or embedding model.Commercial use: allowed; Jakarta hosting: unknown; Indonesian support: listed by developer
SEA-LION ModernBERT 300MAI SingaporeReleasedOct 2026Runs onCPU serverClassembedding
AI Singapore SEA-LION ModernBERT 300M encoder or embedding model.Commercial use: allowed; Jakarta hosting: unknown; Indonesian support: listed by developer
SEA-LION ModernBERT 600MAI SingaporeReleasedOct 2026Runs onCPU serverClassembeddingAI Singapore SEA-LION ModernBERT 600M encoder or embedding model.Commercial use: allowed; Jakarta hosting: unknown; Indonesian support: listed by developer
multilingual-e5-smallMicrosoft AIReleasedJun 2023Runs onnot in the notesClassembeddingSmall multilingual E5 text embedding model for retrieval and similarity across about 100 languages.Commercial use: allowed; Jakarta hosting: unknown; Indonesian support: listed by developerWhere this comes from
Models the research names for this use case. Research, with sources on each model record.
Who Runs It
No deployment is recorded for this use case.
Where this comes from
Documented uses by named organisations. The evidence column says who made the claim.
Candidate Models
MMTEB
A general benchmark for this model class, not a test on this use case. score. Higher is better.
Nemotron-3-Embed-8BNVIDIA vendor-reported75.45050100
EmbeddingGemma 2Google, embeddingReleasedOct 2026open weightshas scoreslast 18 months
Nemotron-3-Embed-8BNVIDIA, embeddingReleasedJul 2026open weightshas scoreslast 18 months
harrier-oss-v1-0.6bMicrosoft AI, embeddingReleasedMar 2026open weightsIndonesian statedlast 18 months
Qwen3-Embedding-8BAlibaba (Qwen), embeddingReleasedJun 2025open weightshas scoreslast 18 months
LFM2.5-Encoder 230MLiquid AI, embeddingReleasedOct 2026open weightslast 18 months
LFM2.5-Encoder 350MLiquid AI, embeddingReleasedOct 2026open weightslast 18 months
Nemotron-3-Embed-1BNVIDIA, embeddingReleasedJul 2026open weightslast 18 months
Granite Embedding Multilingual R2IBM, embeddingReleasedMay 2026open weightslast 18 monthsShowing 10 of 115. All models of this class
Where this comes from
Derived, not researched: models in the directory whose class matches a class this use case uses, excluding family records and models shut down or deprecated. Ranked by how many fit facts they meet, then by release date. Each fit fact is read from the model record.
Links
Sources
- aisingapore (AI Singapore) huggingface.co https://huggingface.co/aisingapore
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Embeddings and Retrieval for SEA Languages", https://aurion.semai.id/usecases/pub-sea-embeddings/, records of October 2026.