Gemma SEA LION v3 9B IT

Gemma

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released April 20258K context≈ 6,144 words

A regionally-focused language model built on Google's Gemma architecture, SEA LION v3 9B IT is tuned specifically for Southeast Asian languages and contexts. It handles the linguistic diversity of the region — covering languages like Indonesian, Thai, Vietnamese, and others — with more cultural grounding than general-purpose models. At 9B parameters with an 8K context window, it trades raw scale for regional specificity.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Multilingual

Strong

Coding

Gemma SEA LION v3 9B IT

Gemma

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released April 20258K context≈ 6,144 words

A regionally-focused language model built on Google's Gemma architecture, SEA LION v3 9B IT is tuned specifically for Southeast Asian languages and contexts. It handles the linguistic diversity of the region — covering languages like Indonesian, Thai, Vietnamese, and others — with more cultural grounding than general-purpose models. At 9B parameters with an 8K context window, it trades raw scale for regional specificity.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Multilingual

Strong

Coding

Glossary

ArchitectureThe underlying structural design of a neural network that defines how data flows through layers and components.Context WindowThe maximum number of tokens a model can process in a single conversation or prompt.Gemma ArchitectureA transformer-based neural network design created by Google that serves as the structural foundation for this model.General-PurposeDesigned to handle a wide variety of different tasks rather than being specialized for one specific domain.GroundingThe practice of ensuring a model's responses are based on and supported by provided source documents rather than generated from general knowledge.Language ModelAn AI model trained to predict and generate text by learning patterns from large amounts of written data.Linguistic DiversityThe existence and coexistence of multiple languages and language varieties within a community or system.ParametersThe learned numerical values in a model — more parameters generally means more capacity but higher compute cost.SpecificityThe proportion of negative cases correctly identified; high specificity means fewer false alarms.

Capabilities

Capabilities

Use Case Fit

Glossary