Router DeskHomeQwen 3.5 9B
Qwen 3.5 9B
Small enough to be cheap on output, new enough to take images. The slug uses a dot: qwen/qwen3.5-9b.
Updated 2026-09-27
Qwen3.5-9B is a 9 billion parameter multimodal model. OpenRouter describes a unified vision-language design aimed at reasoning, coding, and visual understanding. On 27 Sep 2026 the price was $0.10 per million input tokens and $0.15 per million output tokens, with a 262,144-token context.
The search spelling “qwen3 5 9b” omits the dot. The id does not. qwen/qwen3-32b is a different model from an earlier release. Qwen3.7 Max is the large flagship, not a 9B.
- Use 9B when the task fits a short answer and you want output near $0.15 per million.
- Move to Qwen3 32B or 235B A22B when the 9B misses a reasoning step you can see in the trace.
- Move to a Gemini or GPT image slug when you need an image back, not a description of an image.
Questions
What is the Qwen 3.5 9B model id?
qwen/qwen3.5-9b
How much does it cost?
$0.10 per million input tokens and $0.15 per million output tokens on 27 Sep 2026.
Does it read images?
The model card describes visual understanding. It returns text. It is not an image generator.
Related
- Qwen3 32B
Qwen3 32B is a dense 32.8B model. On 27 Sep 2026 OpenRouter priced qwen/qwen3-32b at $0.08 input and $0.28 output per million tokens.
- Qwen3 Max
Qwen3.7 Max was $1.475 input and $4.425 output per million tokens on 27 Sep 2026, with a 1M context. The older qwen/qwen3-max slug was $0.78 / $3.90.
- Qwen3 14B
Qwen3 14B is a dense 14.8B model. OpenRouter listed qwen/qwen3-14b at $0.12 input and $0.24 output per million tokens.
- Image router
OpenRouter’s image collection includes Gemini and GPT image slugs. Video such as Seedance is billed per second, not as a chat completion.