Qwen3.8 27B Brings Open-Weight Multimodal Power to Poly
Qwen3.8 27B is now available on Poly through OpenRouter. It is a dense, open-weight vision-language model aimed at coding, research, professional workflows, multimodal interaction, and long-running agents.
At 27 billion parameters, it sits in an increasingly useful middle ground: large enough for serious work, but much more approachable than trillion-parameter frontier systems. The open-weight release also gives teams a path to study, adapt, or self-host the model outside managed APIs when their requirements call for it.
Multimodal and Agent Ready
Qwen3.8 27B accepts text, images, and video and returns text. Through OpenRouter it supports tool calling, structured outputs, long responses, and configurable reasoning at low, medium, and extra-high effort.
Good use cases include:
- Coding and codebase analysis with tool access
- Visual debugging from screenshots or recordings
- Research across long documents and mixed media
- Structured extraction and professional knowledge workflows
- Agent tasks that benefit from deliberate reasoning without flagship pricing
Context and Pricing
OpenRouter lists a 262,144-token context window with up to 131,072 output tokens. Current pricing is $0.45 per million input tokens and $3.20 per million output tokens.
That pricing makes Qwen3.8 27B an interesting alternative when a workflow needs vision, reasoning, and tools but does not justify Qwen3.8 Max for every step.
27B or Max?
Use Qwen3.8 27B for the main loop, especially when cost, openness, or deployment flexibility matters. Move to Qwen3.8 Max when you need the flagship model's one-million-token context or want the strongest available Qwen pass for a difficult final stage.
On Poly, you can make that switch inside the same conversation and compare the result without rebuilding the prompt.
Try Qwen3.8 27B Compare Qwen Models
Source: Qwen3.8 27B on OpenRouter.