LLMfamily
Qwen3
Alibaba's Qwen3 family (base for several derivatives in the notes).
Levelfamily
ClassLLM
Parametersnot in the notes
Contextnot in the notes
Released29 Apr 2025
Opennessopen weights permissive
Runs onnot in the notes
Statusactive
Benchmark Results
maintainer-run
SEA-HELMSoutheast Asian and Indonesian
Qwen3-8BCheckpoint Qwen3-8B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
69.36 95% CI [-1.32, +1.36]
SEA-HELM leaderboard, 58 models, Indonesian (ID)
Qwen3-8BCheckpoint Qwen3-8B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
60.33 95% CI [-0.66, +0.67]
04080
Score in score. Higher is better.
SEA-HELM leaderboard, 58 models, SEA average
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
finetunecontinued pretrain
Qwen3-ASR is a child (finetune)low confidencenotesQwen3-Embedding-0.6B is a child (finetune)medium confidencenotesQwen3-Embedding-4B is a child (finetune)medium confidencenotesQwen3-Embedding-8B is a child (finetune)medium confidencenotesQwen3-Reranker-0.6B is a child (finetune)medium confidencenotesQwen3-Reranker-4B is a child (finetune)medium confidencenotesQwen3-Reranker-8B is a child (finetune)medium confidencenotesQwen3-VL-Embedding-2B is a child (finetune)medium confidencenotesQwen3-VL-Embedding-8B is a child (finetune)medium confidencenotes
Qwen3-Swallow is a child (continued pretrain)medium confidencenotesQwen3-0.6B-Base belongs to this familyLLM28 Apr 2025600MQwen3-4B-Base belongs to this familyLLM28 Apr 2025Qwen3-0.6B belongs to this familyLLM27 Apr 2025751.63MQwen3-1.7B belongs to this familyLLM27 Apr 20251.7BQwen3-235B-A22B belongs to this familyLLM27 Apr 2025235.09BQwen3-30B-A3B belongs to this familyLLM27 Apr 202530.53BQwen3-32B belongs to this familyLLM27 Apr 202532.76BQwen3-4B belongs to this familyLLM27 Apr 20254BQwen3-8B belongs to this familyLLM27 Apr 20258BWhere 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, 11 models. Confidence and evidence are the notes' own.
R&D Topics
Identity
- Developer
- Alibaba (Qwen)
- Family
- not in the notes
- Released
- 29 Apr 2025 sourcesnotes
- 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
- other dense and MoE
- Total parameters
- – sources
- Active parameters
- –
- Checkpoint parameters
- –
- Note
- 0.6B to 235B; small models distilled from flagship knowledge.
- 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
- not stated
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- 119 languages and dialects (up from 29)
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
DeveloperAlibaba (Qwen)big tech, China
Alibaba (Qwen)big tech, ChinaModelQwen3Alibaba (Qwen), LLM
Qwen3Alibaba (Qwen), LLMR&D TopicsDistillation of Frontier Models into Smaller OnesOpen-Weight Frontier Releases as Research SubstrateSmall Language Models (SLM)
Use CasesIndonesian and Regional-Language LLMsPublic SectorTelecom Domain LLMs and BenchmarksTelecomIndustrial Copilots and Maintenance AssistantsIndustrialClinical LLM (Open Weights)Healthcare+1 moreUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs It
PT Pertamina Geothermal EnergyIn production, 14 Jul 2026
PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026
Siemens Erlangen electronics factoryPlanned, 6 Jan 2026MIND IDResearch, 14 Oct 2025+16 moreDeployments of these use cases; none names this model.
PT Pertamina Geothermal EnergyIn production, 14 Jul 2026PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026Siemens Erlangen electronics factoryPlanned, 6 Jan 2026MIND IDResearch, 14 Oct 2025+16 moreDeployments of these use cases; none names this model.Models 19
Qwen3-8BAlibaba (Qwen), LLMFamilyQwen3-0.6BAlibaba (Qwen), LLMFamilyQwen3-0.6B-BaseAlibaba (Qwen), LLMFamilyQwen3-4BAlibaba (Qwen), LLMFamilyQwen3-ASRAlibaba (Qwen), speech recognitionfinetuneQwen3-1.7BAlibaba (Qwen), LLMFamilyQwen3-235B-A22BAlibaba (Qwen), LLMFamilyQwen3-30B-A3BAlibaba (Qwen), LLMFamilyQwen3-4B-BaseAlibaba (Qwen), LLMFamilyQwen3-VL-Embedding-2BAlibaba (Qwen), embeddingfinetuneQwen3-VL-Embedding-8BAlibaba (Qwen), embeddingfinetuneQwen3-32BAlibaba (Qwen), LLMFamily+7
Qwen3-Embedding-0.6BAlibaba (Qwen), embeddingfinetuneQwen3-Embedding-4BAlibaba (Qwen), embeddingfinetuneQwen3-Embedding-8BAlibaba (Qwen), embeddingfinetuneQwen3-Reranker-0.6BAlibaba (Qwen), rerankerfinetuneQwen3-Reranker-4BAlibaba (Qwen), rerankerfinetuneQwen3-Reranker-8BAlibaba (Qwen), rerankerfinetuneQwen3-SwallowInstitute of Science Tokyo, LLMcontinued_pretrainOrganisations 1
R&D topics 3
Open-Weight Frontier Releases as Research SubstrateUsing released weights as the base for downstream research.exemplarSmall Language Models (SLM)Models under about 10B parameters that fit consumer devices.exemplarDistillation of Frontier Models into Smaller OnesTransferring capability from a large teacher model to a smaller one.exemplar
Sources
- tokyotech-llm (tokyotech-llm) huggingface.co https://huggingface.co/tokyotech-llm
- Qwen/Qwen3Guard-Stream-4B , Hugging Face huggingface.co https://huggingface.co/Qwen/Qwen3Guard-Stream-4B
- Gladia - Best open-source speech-to-text models in 2026 gladia.io https://www.gladia.io/blog/best-open-source-speech-to-text-models
- Qwen 3 offers a case study in how to effectively release a model simonwillison.net https://simonwillison.net/2025/Apr/29/qwen-3/
- huggingface.co/api/models/Qwen/Qwen3-0.6B https://huggingface.co/api/models/Qwen/Qwen3-0.6B
- Qwen3 Technical Report arxiv.org https://arxiv.org/abs/2505.09388
- qwenlm.github.io/blog/qwen3 https://qwenlm.github.io/blog/qwen3/
- huggingface.co/Qwen/Qwen3-4B https://huggingface.co/Qwen/Qwen3-4B
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Qwen3", https://aurion.semai.id/models/alibaba-qwen3/, records of October 2026.