Qwen: Qwen3.5 397B A17B
Qwen3.5 397B A17B is capable of processing text, images, and videos, with a context length of 262,144 tokens and the ability to support reasoning and tools for complex tasks. It offers structured output options but does not provide open-source weights, meaning it cannot be used freely or modified by developers. This model stands as a strong contender for those requiring robust multimodal handling at a price point of $0.39 per million input tokens and $2.34 per million output tokens. With a blended benchmark score of 63.2 across four independent evaluations, it excels in specific areas like coding but has less proven performance in agentic tasks. Its suitability for users who need balanced multimodal capabilities and are willing to pay the premium can be highly recommended.
Benchmark results
Independent, published benchmarks. Blended score 63.2 across 4 benchmarks, last refreshed 2026-07-31. How scoring works →
| Benchmark | Measures | Score |
|---|---|---|
| AI Index | broad capability composite | 55.6 |
| AI Index Coding | software engineering tasks | 79.5 |
| AI Index Agentic | multi-step tool-using tasks | 32.7 |
| EQ-Bench | emotional understanding in dialogue | 71.3 |
- Model ID
- qwen/qwen3.5-397b-a17b
- Vendor
- qwen
- Released
- February 2026
- Tokenizer
- Qwen3
- Input Modalities
- text, image, video
- Output Modalities
- text
- Max Output
- 65,536 tokens
- Tool Calling
- ✓ supported
- Structured Output
- ✓ supported
- Reasoning Mode
- ✓ supported
- Vision
- ✓ accepts images
- Audio
- no
- Moderated
- no
What it costs in practice
Computed from the current $0.39/M input and $2.34/M output rates. Run your own numbers →
| Job | Tokens | Cost |
|---|---|---|
| Summarize a 50-page report | 30k in / 1.5k out | $0.02 |
| Classify 1,000 customer emails | 500k in / 50k out | $0.31 |
| A month of a busy support chatbot | 5M in / 2M out | $6.63 |
Category rankings
Where Qwen: Qwen3.5 397B A17B places across the 10 categories it ranks in. How we rank →
| # | Category | Score |
|---|---|---|
| #11 | Video SummarizationVideo · of 25 ranked | 153 |
| #12 | Short-Form SummarizationWriting · of 25 ranked | 130 |
| #12 | Email DraftingWriting · of 25 ranked | 126 |
| #12 | Language LearningEducation · of 25 ranked | 126 |
| #12 | Chat CompanionPersonal · of 25 ranked | 130 |
| #13 | Trivia & General KnowledgePersonal · of 25 ranked | 120 |
| #21 | Image CaptioningVision · of 25 ranked | 120 |
| #22 | Dataset AnnotationResearch · of 25 ranked | 141 |
| #24 | JSON ExtractionData · of 25 ranked | 144 |
| #25 | Bulk Data LabelingData · of 25 ranked | 133 |
Similar models
Qwen: Qwen3.7 Flash
Qwen: Qwen3.7 Plus
Qwen: Qwen3.5 Plus 2026-04-20
Qwen: Qwen3.6 Flash
Qwen: Qwen3.6 35B A3B
Qwen: Qwen3.6 27B
Quick answers
- How much does Qwen: Qwen3.5 397B A17B cost?
- $0.39 per million input tokens and $2.34 per million output tokens.
- What is Qwen: Qwen3.5 397B A17B's context window?
- 262,144 tokens, roughly 393 pages of text in a single request.
- Does Qwen: Qwen3.5 397B A17B support tool calling?
- It supports tool calling, structured output, a reasoning mode.
- Can Qwen: Qwen3.5 397B A17B process images?
- Yes, it accepts image input.