multimodal LLM
Gemma 4 31B
Google's 31B dense text and image model, top open model on SEA-HELM (18 Sep 2026); tech report 2 Jul 2026. 31B-it: finetunes 287, adapters 331, merges 38, quants 325.
Levelmodel
Classmultimodal LLM
Parameters30.7B
Context256,000
Released2 Apr 2026
Opennessopen weights permissive
Runs onsingle datacentre GPU, consumer GPU
Statusactive
Benchmark Results
vendor-reportedmaintainer-run
GPQAknowledge and reasoning
Reported as 'GPQA' by secondary source; subset not statedvendor-reported
84.3
050100
Score in %. Higher is better.
GPQA Diamondknowledge and reasoning
vendor-reported
84.3
050100
Score in %. Higher is better.
MMLU-Proknowledge and reasoning
vendor-reported
85.2
050100
Score in %. Higher is better.
SEA-HELMSoutheast Asian and Indonesian
Indonesianmaintainer-run
79.75 95% CI [-1.25, +1.20]
04080
Score in index. Higher is better.
rank 1, SEA-HELM Indonesian
maintainer-run
79.75 95% CI [-1.25, +1.20]
04080
Score in score. Higher is better.
rank 1, SEA-HELM Indonesian, open-weight models, average of 2,000 bootstraps
maintainer-run
76.17 95% CI [-0.60, +0.59]
04080
Score in score. Higher is better.
rank 1, SEA-HELM SEA average, open-weight models
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-SEA-LION-v4.5-E2B-IT is a child (distill)high confidencemodel cardWhere 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.
Identity
- Developer
- Google, Google DeepMind
- Family
- Gemma 4
- Released
- 2 Apr 2026 sourcesannouncement page
- 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
- dense transformer Dense; hybrid local sliding-window + global attention, p-RoPE
- Total parameters
- 30.7B sources
- Active parameters
- –
- Checkpoint parameters
- –
- Note
- 31B (flagship of the family)
- Context length
- 256,000 tokens
- Input
- text, image
- 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
- single datacentre GPU, consumer GPU
- Footprint
- Single datacentre GPU; 4-bit on consumer GPU
- 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): 79.75 (95% CI -1.25 to +1.20), run by AI Singapore (SEA-HELM leaderboard maintainer) leaderboard.sea-lion.ai
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Industry Use Cases
No industry use case links this model.
Links
ModelGemma 4 31BGoogle, multimodal LLM
Gemma 4 31BGoogle, multimodal LLMUse CasesClinical LLM (Open Weights)HealthcareProcess Optimisation and Advanced Process ControlIndustrialTelecom Domain LLMs and BenchmarksTelecomUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs It
PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026MIND IDResearch, 14 Oct 2025
Conch CementIn production, 23 Jun 2025
PT Kalbe Morinaga IndonesiaIn production, Oct 2024+2 moreDeployments of these use cases; none names this model.
PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026MIND IDResearch, 14 Oct 2025Conch CementIn production, 23 Jun 2025PT Kalbe Morinaga IndonesiaIn production, Oct 2024+2 moreDeployments of these use cases; none names this model.Models 2
Sources
- google/gemma-4-31B-it , Hugging Face huggingface.co https://huggingface.co/google/gemma-4-31B-it
- google (Google) huggingface.co https://huggingface.co/google
- SEA-HELM Leaderboard leaderboard.sea-lion.ai/detailed/SEA https://leaderboard.sea-lion.ai/detailed/SEA
- SEA-HELM Leaderboard leaderboard.sea-lion.ai/detailed/ID https://leaderboard.sea-lion.ai/detailed/ID
- Gemma 4: Byte for byte, the most capable open models blog.google https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/
- huggingface.co/api/models/google/gemma-4-31B-it/commits/main https://huggingface.co/api/models/google/gemma-4-31B-it/commits/main
- Google Releases Gemma 4: Open Models Under Apache 2.0 datanorth.ai https://datanorth.ai/news/google-releases-gemma-4-open-models
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Gemma 4 31B", https://aurion.semai.id/models/google-gemma-4-31b/, records of October 2026.