Project AURIONAI Use-Cases Research & Innovation
Alibaba (Qwen)
multimodal LLM

Qwen3.5-9B

Qwen3.5 9B small dense model.

Alibaba (Qwen)Alibaba (Qwen)ChinaQwen3.5Family
Benchmark score chart of the small Qwen3.5 models from the model card.
Levelvariant
Classmultimodal LLM
Parameters9B
Context262,000
Released2 Mar 2026
Opennessopen weights permissive
Runs onconsumer GPU
Statusactive

Benchmark Results

vendor-reportedmaintainer-run
GPQA Diamondknowledge and reasoning
Not independently verifiedvendor-reported
81.7
050100

Score in %. Higher is better.

Mar 2026 rits.shanghai.nyu.edu
MMLU-Proknowledge and reasoning
Not independently verifiedvendor-reported
82.5
050100

Score in %. Higher is better.

Mar 2026 rits.shanghai.nyu.edu
SEA-HELMSoutheast Asian and Indonesian
Checkpoint Qwen3.5-9B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
74.23 95% CI [-1.29, +1.28]
04080

Score in score. Higher is better.

SEA-HELM leaderboard, 58 models, Indonesian (ID)18 Sep 2026 leaderboard.sea-lion.ai
Checkpoint Qwen3.5-9B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
69.13 95% CI [-0.64, +0.63]
04080

Score in score. Higher is better.

SEA-HELM leaderboard, 58 models, SEA average18 Sep 2026 leaderboard.sea-lion.ai
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

Small language models (SLM)produced by

Identity

Developer
Alibaba (Qwen)
Family
Qwen3.5
Released
2 Mar 2026 sourcesregistry date
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 Dense; hybrid Gated DeltaNet plus full attention
Total parameters
9B sources
Active parameters
–
Checkpoint parameters
–
Note
9B dense
Context length
262,000 tokens
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
consumer GPU
Footprint
About 18 GB unquantized (article estimate); RTX 3090/4090; 4-bit cuts VRAM about 75%
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.

Sources

  1. Qwen3.5 - a Qwen Collection huggingface.co https://huggingface.co/collections/Qwen/qwen35
  2. Qwen 3.5 Small Models: 9B Parameters That Beat 120B rits.shanghai.nyu.edu https://rits.shanghai.nyu.edu/ai/qwen-3-5-small-models-9b-parameters-that-beat-120b
  3. huggingface.co/Qwen/Qwen3.5-0.8B https://huggingface.co/Qwen/Qwen3.5-0.8B

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Qwen3.5-9B", https://aurion.semai.id/models/alibaba-qwen3-5-9b/, records of October 2026.