Unverified50% confidenceOpinionExact time
SGLang's high-frequency release cadence is aimed at rapidly tracking new version releases of mainstream model families such as Llama, Qwen, and Gemma, as well as new quantization technologies.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/18/2026
First Seen
Valid until: 12/17/2026
Sources
Related Entities
Related Claims
UnverifiedGQA has been adopted by Llama 2 70B and the Llama 3 series, while MQA has been adopted by the Falcon series; both reduce the number of cached KV vectors by having multiple query heads share the same set of KV heads.62% similarUnverifiedModels newly supported in SGLang v0.5.18 include Muse Glimmer, Intern-S2-Mobius, SANA-Video, LingBot-Video-MoE, LTX-2.5, Cosmos3 Edge & Distilled, and LongCat-Image.61% similarUnverified寒武纪已完成GLM、DeepSeek、Qwen、Kimi、MiniMax五款国产主流大模型的推理适配，第六代芯片架构正在研发中61% similarUnverifiedIn the past two weeks, three new models from China have emerged: Kimi K3, DeepSeek's new Flash model, and Qwen 3.8 Max.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/931679API
curl https://kongchang.com/api/v1/knowledge/claims/931679MCP
get_claim(id=931679)