Qwen3.8-27B
Sovereignty and hosting
This model is catalogued for reference. It is not hosted on LLM EU yet, and no first-party endpoint serves it here. Requests to it go to the provider directly, under that provider's own terms.
Summary
Qwen3.8-27B is a dense vision-language model with a 262,144-token context that the vendor says can be extended to one million tokens. Thinking mode is enabled by default and can be disabled per request, with reasoning depth controlled by a reasoning_effort parameter. It is published under Apache-2.0 and is the smallest model of the Qwen3.8 generation.
Specifications
- Provider
- Alibaba Qwen (CN)
- Context window
- 262k
- Parameters
- 27.78B
- Modalities
- text, vision
- License
- apache-2.0 · licence
- Open weights
- yes
- Release date
- 05 Aug 2026
- Tasks
- reasoning, coding, vision, multilingual-support, summarization
Languages
Strong: en, zh
Adequate: de, fr, es, it, pt, ru, ja, ko, ar
Languages
Strong
- en
- zh
Adequate
- de
- fr
- es
- it
- pt
- ru
- ja
- ko
- ar
Not verified: the provider has not documented this claim. — a language is listed as strong only where a source claims it; the card links the evidence where one exists.
When to use it
Strengths
- 262k-token context, extensible to 1M per the vendor
- native image and video understanding
- switchable thinking mode with adjustable reasoning effort
- Apache-2.0 licence
Limits
- benchmark claims on the card are self-reported
- dense 27B weights need a large-memory accelerator at BF16
- no evaluation has been run by LLM EU