multimodal LLM
Qwen3.8-Flash-Next
Alibaba's mid-size MoE model for image, video and text (1M context with YaRN). Adapters 1, finetunes 71, merges 3, quants 391.
Levelmodel
Classmultimodal LLM
Parameters125B
Context262,144
Released24 Aug 2026
Opennessopen weights restricted
Runs onsingle datacentre GPU
Statusactive
Benchmark Results
vendor-reportedmaintainer-run
GPQA Diamondknowledge and reasoning
vendor-reported
91.7
050100
Score in %. Higher is better.
LiveCodeBench v6coding
vendor-reported
91.9
050100
Score in %. Higher is better.
SEA-HELMSoutheast Asian and Indonesian
Checkpoint Qwen3.8-Flash-Next as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
79.04 95% CI [-1.23, +1.16]
04080
Score in score. Higher is better.
SEA-HELM leaderboard, 58 models, Indonesian (ID)
Checkpoint Qwen3.8-Flash-Next as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
73.04 95% CI [-0.53, +0.53]
04080
Score in score. Higher is better.
SEA-HELM leaderboard, 58 models, SEA average
SWE-bench Procoding
vendor-reported
62.5
050100
Score in %. Higher is better.
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).
R&D Topics
Identity
- Developer
- Alibaba (Qwen)
- Family
- Qwen3.8
- Released
- 24 Aug 2026 sourcessecondary report
- 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
- hybrid Gated DeltaNet + Qwen Sparse Attention, MoE (512 experts, 10 + 1 active), n-gram embedding
- Total parameters
- 125B sources
- Active parameters
- 6B
- Checkpoint parameters
- 180B
- Note
- 125B core; 180B including 51B n-gram embedding and 4B MTP
- Context length
- 262,144 tokens
- Input
- text, image, video
- 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 restricted
- Licence
- Qwen Community License 1.0 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
- Footprint
- Single GPU at FP8
- 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
- not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Lineage
The notes record no lineage for this model.
Industry Use Cases
No industry use case links this model.
Links
DeveloperAlibaba (Qwen)big tech, China
Alibaba (Qwen)big tech, ChinaModelQwen3.8-Flash-NextAlibaba (Qwen), multimodal LLM
Qwen3.8-Flash-NextAlibaba (Qwen), multimodal LLMR&D TopicsEfficiency Architectures (MoE, Sparse Attention, Long Context)Frontier Pretraining ScalingOpen-Weight Frontier Releases as Research Substrate
Use CasesIndustrial Copilots and Maintenance AssistantsIndustrialClinical LLM (Frontier Proprietary)HealthcarePharma Documents, Regulatory and Life-Science R&DHealthcareAmbient Clinical ScribingHealthcare+5 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
Banner HealthIn production, 11 Jan 2026CommureIn production, 11 Jan 2026+45 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 2026Banner HealthIn production, 11 Jan 2026CommureIn production, 11 Jan 2026+45 moreDeployments of these use cases; none names this model.Models 1
Organisations 1
R&D topics 3
Open-Weight Frontier Releases as Research SubstrateUsing released weights as the base for downstream research.used byFrontier Pretraining ScalingTraining a new base model at or near the top-5 compute of its release date (Epoch's front…produced byEfficiency Architectures (MoE, Sparse Attention, Long Context)Reducing training and inference cost per capability.produced by
Sources
- Open-Source LLMs 2026: Which Are Actually Open-Licensed codersera.com https://codersera.com/blog/open-source-llms-landscape-2026/
- Qwen (Qwen) huggingface.co https://huggingface.co/Qwen
- Qwen/Qwen3.8-Flash-Next , Hugging Face huggingface.co https://huggingface.co/Qwen/Qwen3.8-Flash-Next
- LICENSE , Qwen/Qwen3.8-Flash-Next at main huggingface.co https://huggingface.co/Qwen/Qwen3.8-Flash-Next/blob/main/LICENSE
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Qwen3.8-Flash-Next", https://aurion.semai.id/models/alibaba-qwen3-8-flash-next/, records of October 2026.