Unverified50% confidenceFactExact time
vLLM v0.27.0 adds support for the Qwen3.5 series (including dense and MoE models) and supports EVS video token pruning.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedSeries such as Qwen2.5 and Gemma3 have already released 1B–3B quantized versions optimized specifically for edge scenarios.70% similarUnverifiedHugging Face community developer llmfan46 released several uncensored fine-tuned models based on the Qwen3 series, including Qwen3.8-27B, Qwen3.5-122B-A10B, Qwen3-Coder-Next, and Laguna-S2.1.69% similarUnverifiedMainstream inference frameworks such as vLLM have initially begun supporting Qwen3.8-Flash-Next, but the feature for offloading Embeddings to CPU memory is still under development.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/898649API
curl https://kongchang.com/api/v1/knowledge/claims/898649MCP
get_claim(id=898649)