Qwen: Qwen3.5-9B (batch)
The Qwen AI model is a versatile tool that supports multiple input modalities: text, image, and video. It can process context up to 262144 tokens and generate output with up to 235929 completion tokens. Qwen may be worth considering for users seeking an AI model with strong reasoning capabilities. Its support for tools and ability to reason make it a good fit for applications requiring more complex interactions. However, its price point is slightly higher than average: $0.17 per million input tokens and $0.25 per million output tokens. Additionally, Qwen's benchmark scores are respectable but not exceptional, with a blended score of 30.8 across three independent benchmarks.
Benchmark results
Independent, published benchmarks. Blended score 30.8 across 3 benchmarks, last refreshed 2026-09-04. How scoring works →
| Benchmark | Measures | Score |
|---|---|---|
| AI Index | broad capability composite | 36.0 |
| AI Index Coding | software engineering tasks | 47.3 |
| AI Index Agentic | multi-step tool-using tasks | 11.5 |
- Model ID
- qwen/qwen3.5-9b:batch
- Vendor
- qwen
- Released
- March 2026
- Tokenizer
- Qwen3
- Input Modalities
- text, image, video
- Output Modalities
- text
- Max Output
- 235,929 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 $0.17/M input and $0.25/M output rates. Run your own numbers →
| Job | Tokens | Cost |
|---|---|---|
| Summarize a 50-page report | 30k in / 1.5k out | under $0.01 |
| Classify 1,000 customer emails | 500k in / 50k out | $0.10 |
| A month of a busy support chatbot | 5M in / 2M out | $1.35 |
Category rankings
Where Qwen: Qwen3.5-9B (batch) places across the 5 categories it ranks in. How we rank →
| # | Category | Score |
|---|---|---|
| #16 | Self-Hosted / LocalCost · of 25 ranked | 117 |
| #16 | Real-Time ChatLatency · of 25 ranked | 118 |
| #19 | Cheap Bulk InferenceCost · of 25 ranked | 137 |
| #24 | Social Media PostsWriting · of 25 ranked | 119 |
| #24 | Voice Assistant BackendVoice · of 25 ranked | 123 |
Similar models
Qwen: Qwen3.8 Flash
Qwen: Qwen3.8 27B
Qwen: Qwen3.8 Max
Qwen: Qwen3.7 Flash
Qwen: Qwen3.7 Plus
Qwen: Qwen3.5 Plus 2026-04-20
Quick answers
- How much does Qwen: Qwen3.5-9B (batch) cost?
- $0.17 per million input tokens and $0.25 per million output tokens.
- What is Qwen: Qwen3.5-9B (batch)'s context window?
- 262,144 tokens, roughly 393 pages of text in a single request.
- Does Qwen: Qwen3.5-9B (batch) support tool calling?
- It supports tool calling, structured output, a reasoning mode.
- Can Qwen: Qwen3.5-9B (batch) process images?
- Yes, it accepts image input.