Project AURIONAI Use-Cases Research & Innovation
AI Singapore
edge LLM

Gemma-SEA-LION-v4.5-E2B-IT

AI Singapore's small edge SEA model from gemma-4-E2B-it, distilled from gemma-4-31B-it plus merging.

AI SingaporeAI SingaporeSingaporeSEA-LIONFamily
SEA-HELM leaderboard chart of Southeast Asian language scores for models under 10B parameters, from the model card.
Levelmodel
Classedge LLM
Parameters5.1B
Contextnot in the notes
Released19 May 2026
Opennessopen weights permissive
Runs onlaptop, mobile or NPU
Statusactive

Benchmark Results

maintainer-run
SEA-HELMSoutheast Asian and Indonesian
maintainer-run
63.96 95% CI [-1.38, +1.37]
04080

Score in score. Higher is better.

SEA-HELM Indonesian, open-weight models, average of 2,000 bootstraps18 Sep 2026 leaderboard.sea-lion.ai
maintainer-run
59.23 95% CI [-0.66, +0.67]
04080

Score in score. Higher is better.

SEA-HELM SEA average, open-weight models18 Sep 2026 leaderboard.sea-lion.ai
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).

Lineage

distillfinetune
Gemma 4 31B to Gemma-SEA-LION-v4.5-E2B-IT: distill, high confidenceGemma 4 E2B to Gemma-SEA-LION-v4.5-E2B-IT: finetune, high confidenceGemma to Gemma-SEA-LION-v4.5-E2B-IT: finetune, low confidenceGemmaGemmaGoogleGemma 4 31BGemma 4 31BGoogle, Apr 2026Gemma 4 E2BGemma 4 E2BGoogle, Apr 2026Gemma-SEA-LION-v4.5-E2B-ITGemma-SEA-LION-v4.5-E2B-ITAI Singapore, May 2026
GoogleGemma 4 31B is the parent (distill)high confidencemodel card
GoogleGemma 4 E2B is the parent (finetune)high confidencemodel card
GoogleGemma is the parent (finetune)low confidencenotes
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 4 models. Confidence and evidence are the notes' own.

Industry Use Cases

Public Sector
Indonesian and regional-language LLMs

R&D Topics

Distillation of frontier models into smaller onesused byLow-resource and regional language resourcesproduced byQuantization and edge inferenceproduced byRegional or national foundation model training (continued pretraining at 7B to 70B)produced bySmall language models (SLM)produced by

Identity

Developer
AI Singapore
Family
SEA-LION
Released
19 May 2026 sourcesregistry date
Weights published
–
Record checked
10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
not in the notes
Total parameters
5.1B sources
Active parameters
2.3B
Checkpoint parameters
–
Note
2.3B effective (5.1B total)
Context length
not in the notes
Input
text
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
MIT License sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
laptop, mobile or NPU
Footprint
GGUF
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
measured on a benchmark
  • SEA-HELM Indonesian leaderboard (ID aggregate, first score column of the Indonesian Competency and Task Scores table): 63.96 (95% CI -1.38 to +1.37), run by AI Singapore (SEA-HELM leaderboard maintainer) leaderboard.sea-lion.ai
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
9 languages including Indonesian
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Sources

  1. aisingapore/Gemma-SEA-LION-v4.5-E2B-IT , Hugging Face huggingface.co https://huggingface.co/aisingapore/Gemma-SEA-LION-v4.5-E2B-IT
  2. SEA-HELM Leaderboard leaderboard.sea-lion.ai/detailed/SEA https://leaderboard.sea-lion.ai/detailed/SEA
  3. SEA-HELM Leaderboard leaderboard.sea-lion.ai/detailed/ID https://leaderboard.sea-lion.ai/detailed/ID
  4. aisingapore (AI Singapore) huggingface.co https://huggingface.co/aisingapore

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Gemma-SEA-LION-v4.5-E2B-IT", https://aurion.semai.id/models/ai-singapore-gemma-sea-lion-v4-5-e2b-it/, records of October 2026.