Expired60% confidenceFactTime unknown
SGLang currently cannot run the Qwen3 27B NVFP4 MTP-XS model, whereas vLLM already supports it.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedThe Qwen3.6 35B model natively supports MTP, and when combined with oMLX, the acceleration capability can be activated directly without any additional configuration of a separate draft model.76% similarUnverifiedQwythos-9B is built on top of the uncensored Qwen 3.5 base model.71% similarUnverifiedThe Qwen3.6 series does not support Fill-in-the-Middle (FIM), and it is recommended to retain Qwen2.5 Coder for real-time code completion.69% similarUnverifiedThe deployment method officially recommended by vLLM does not support running FP8 models on Turing architecture cards such as the 2080 Ti.67% similarUnverifiedThe Qwen3 series also launched six dense models ranging from 0.6B to 32B, all released under the Apache 2.0 license.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/18027API
curl https://kongchang.com/api/v1/knowledge/claims/18027MCP
get_claim(id=18027)