Qwen3.5 SmallOpen Weight

Released

Qwen3.5 Small is an AI model released by Qwen on Monday, Mar 2 2026, 6 days after Qwen3.5 Medium. It is an open-weight model — the trained weights are available to download and run. It was released in 0.8B and 2B and 4B and 9B parameter sizesThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality., with a 256k token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover LiveCodeBench, IFEval, GPQA Diamond, IFBench, MMLU-Pro, OSWorld-Verified, and 3 more.

Benchmarks

Coding

LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better.
65.6%
#44 of 86

Agentic & tool use

IFEvalAgentic tasks — following verifiable instructions in generated responses
91.5%
#3 of 24
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.
41.8%
#35 of 36

Reasoning & science

GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better.
81.7%
#63 of 145
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better.
64.5%
#12 of 16

Knowledge work

MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark
82.5%
#21 of 58

Multimodal

CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better.
73%
#27 of 32
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better.
70.1%
#29 of 36
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better.
78.4%
#10 of 50

Compare Qwen3.5 Small with

Qwen3.5 Small

About

The Qwen3.5 Small release shipped on March 2, 2026 with 0.8B, 2B, 4B and 9B open-weight models. It completed the Qwen3.5 size lineup and offered thinking and visual understanding at sizes intended for local and edge deployments.

The benchmark scores on this page come from the 9B checkpoint, which scored 81.7 on GPQA Diamond and 65.6 on LiveCodeBench v6. The smaller checkpoints offered different memory and performance trade-offs under the same family name.

Frequently asked questions

Qwen3.5 Small was released by Qwen on Monday, Mar 2 2026.

All Qwen releases

42 tracked