Unverified50% confidenceFactExact time
HyperQwen is an Apache-2.0-licensed custom vLLM build released by GitHub user syv-ai, tuned for the Qwen3.8-27B model in the W4A16 quantisation format, with weights of about 19.5GB and DFlash2 speculative decoding.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Related Entities
Related Claims
UnverifiedQwen3.8 Max预览版已上线API，可通过国内百炼/Token Plan或海外qwen.aliyun.com使用，并兼容OpenAI和Anthropic协议端点64% similarUnverifiedDevelopers can access Qwen3.8-Max through three channels: the official blog, Qwen Studio, and the API.63% similarUnverifiedQwen3 27B is licensed under Apache 2.0 and supports image and video input.63% similarUnverifiedChinese models such as DeepSeek and Qwen mostly adopt Apache 2.0 or similarly permissive licenses, allowing developers worldwide to freely download weights, deploy locally, and use them commercially.62% similarVerifiedQwen (Tongyi Qianwen) is a large language model series developed by Alibaba DAMO Academy; it has been continuously open-sourced and iterated since 2023 under the Apache 2.0 license.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1043368API
curl https://kongchang.com/api/v1/knowledge/claims/1043368MCP
get_claim(id=1043368)