Qwen 3.5 35B A3B
Language modelActive
Qwen 3.5 35B A3B
Alibaba
- Released
- February 24, 2026
- Data date
- September 8, 2026
- Speed
- 147.6 tokens/sOutput throughput
- Price
- Input · Output / 1M tokens
- Input
- Text, Image, Video
- Output
- Text
Qwen 3.5 35B A3B brings the multimodal Qwen3.5 architecture into a much smaller MoE configuration. Three billion of its 35 billion parameters are active per token. You can run the checkpoint in Thinking mode or use the instruct configuration for direct responses without visible reasoning.
The model card lists 262,144 native context tokens, with a possible extension to 1,010,000. The weights use Apache-2.0.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | Open weightsSource |
| Access | API, LocalSource |
| Parameters | 35BSource |
| Active parameters | 3BSource |
| Context window | 262,144 tokensSource |
| Weights repository | Qwen/Qwen3.5-35B-A3BSource |
| License | Apache-2.0Source |
| Architecture | hybrid Gated DeltaNet + Gated Attention; dense or sparse MoE with vision encoderSource |
| Input | Text, Image, VideoSource |
| Output | TextSource |
| Reasoning mode | thinking by default; switchable non-thinking/instruct modeSource |
| Tool calling | YesSource |
| Structured output | YesSource |
| Native context window | 262,144 tokensSource |
| Maximum context window | 1,010,000 tokensSource |
| Parameters | 35 BSource |
| Active parameters | 3 BSource |
| Languages | 201 languages and dialectsSource |
| Repository created | 2026-02-24 (Repository creation date is a precise publication proxy, but may differ from the public announcement date or an individual checkpoint's first availability.)Source |
Page 1 of 8
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
MMLU-Pro (Show measurement, test conditions, and source)
- Source value
- 85.3
- Score
- 85.3
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
MMLU-Redux (Show measurement, test conditions, and source)
- Source value
- 93.3
- Score
- 93.3
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
C-Eval (Show measurement, test conditions, and source)
- Source value
- 90.2
- Score
- 90.2
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
SuperGPQA (Show measurement, test conditions, and source)
- Source value
- 63.4
- Score
- 63.4
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
IFEval (Show measurement, test conditions, and source)
- Source value
- 91.9
- Score
- 91.9
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
IFBench (Show measurement, test conditions, and source)
- Source value
- 70.2
- Score
- 70.2
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
MultiChallenge (Show measurement, test conditions, and source)
- Source value
- 60
- Score
- 60
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
AA-LCR (Show measurement, test conditions, and source)
- Source value
- 58.5
- Score
- 58.5
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
LongBench v2 (Show measurement, test conditions, and source)
- Source value
- 59
- Score
- 59
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
HLE w/ CoT (Show measurement, test conditions, and source)
- Source value
- 22.4
- Score
- 22.4
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
GPQA Diamond (Show measurement, test conditions, and source)
- Source value
- 84.2
- Score
- 84.2
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
HMMT Feb 25 (Show measurement, test conditions, and source)
- Source value
- 89
- Score
- 89
- Metric
- percent
- Unit
- %
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Qwen official Hugging Face model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Official model-card comparison table; exact target column, setup as published
Page 1 of 8
Our tests
SVG short test
One SVG prompt and fixed output settings. Captures are shown only with 0 reported reasoning tokens or documented lack of reasoning controls. Identical test conditions do not make the models' internal compute identical.
Earlier tests
Pelican on a bicycle
A community classic for free-form SVG drawing.
Qwen 3.5 35B A3B
Reasoning (requested): high
Current attempt
- Run date
- Oct 1, 2026
- API provider
- openrouter
- Requested API model
- qwen/qwen3.5-35b-a3b
- API model reported in response
- qwen/qwen3.5-35b-a3b
- Reasoning setting
- high
- Pinned endpoint
- Alibaba | qwen/qwen3.5-35b-a3b-20260224
- Temperature
- Provider default
- Token limit including reasoning
- 8,192 tokens
- Test protocol
- community-visual-tests-v1
- Technical checks
- Passed, not a quality rating
- Response time
- 24.1 s
Visually reviewed. This individual test does not establish overall model quality.
Prompt and test conditions
Generate an SVG of a pelican riding a bicycle
Token limit including reasoning: 8,192 tokens
Task origin (Simon Willison)Popularity
42.9 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Speed | Alibaba. Output speed after the first streaming chunk, not total response time. Classified as deprecated by Artificial Analysis. Artificial Analysis maintains only the default 10K-input workload there. Retrieved September 8, 2026. |
| Price | Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. |
| Input | Documented modalities. Retrieved September 8, 2026. |
| Output | Documented modalities. Retrieved September 8, 2026. |
| Research status | Research date September 8, 2026. 3 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not found in the inspected sources | knowledge-cutoff Original wording No exact training-data cutoff was stated in the checked official card/report. |
| Additional source | Model metadata (retrieved September 8, 2026) |
| Additional source | Alibaba Cloud Model Studio visual understanding · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Creative Commons Attribution 4.0 |