translation
MADLAD-400 3B MT
MADLAD-400 3B MT, a checkpoint of the MADLAD-400 MT family. T5-based translation models trained on the MADLAD-400 web corpus, translating across more than 400 languages.
Levelvariant
Classtranslation
Parameters2.94B
Contextnot in the notes
Released27 Nov 2023
Opennessopen weights permissive
Runs onnot in the notes
Statusactive
Industry Use Cases
Public Sector
Translation including IndonesianR&D Topics
Identity
- Developer
- Family
- MADLAD-400 MT
- Released
- 27 Nov 2023 sourcesregistry date
- Weights published
- –
- Record checked
- 12 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
- dense transformer
- Total parameters
- 2.94B sources
- Active parameters
- –
- Checkpoint parameters
- –
- Note
- Parameter count from the safetensors index on Hugging Face.
- 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
- Apache License 2.0 sources
- Access routes
- open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.
Deployment
- Runs on
- not in the notes
- Footprint
- not in the notes
- 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
- listed by developer
- Regional languages
- Vietnamese, Thai, Malay, Burmese, Khmer, Lao, Javanese, Sundanese, Minangkabau, Acehnese, Balinese, Madurese
- Jakarta hosting
- unknown
- Languages
- Model card lists 419 languages, including Indonesian
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Benchmark Results
The notes record no benchmark result for this model.
Lineage
The notes record no lineage for this model.
Links
DeveloperGooglebig tech, USA
Googlebig tech, USAModelMADLAD-400 3B MTGoogle, translation
MADLAD-400 3B MTGoogle, translationWho Runs It
BRIN PRSDI with Universitas Hasanuddin and Universitas HandayaniResearch, 28 Jun 2025Google TranslateIn production, 27 Jun 2024Deployments of these use cases; none names this model.
BRIN PRSDI with Universitas Hasanuddin and Universitas HandayaniResearch, 28 Jun 2025Google TranslateIn production, 27 Jun 2024Deployments of these use cases; none names this model.Organisations 1
Sources
- arxiv.org/abs/2309.04662 https://arxiv.org/abs/2309.04662
- huggingface.co/google/madlad400-3b-mt https://huggingface.co/google/madlad400-3b-mt
- huggingface.co/api/models?pipeline_tag=translation&sort=like... https://huggingface.co/api/models?pipeline_tag=translation&sort=likes&limit=50
- ai.azure.com/catalog/models?task=object-detection https://ai.azure.com/catalog/models?task=object-detection
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "MADLAD-400 3B MT", https://aurion.semai.id/models/google-madlad400-3b-mt/, records of October 2026.