Expired50% confidenceBenchmarkExact time
Through allreduce fusion optimization, the Qwen3-Next model achieved a 71.4% improvement in throughput and reduced Time to First Token (TTFT) from 456ms to 168ms.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/11/2026
First Seen
Valid until: 10/9/2026(expired)
Sources
AI智能体辅助SGLang内核优化：实测性能提升与工程方法论
twitterlmsysorg7/2/2026
Related Claims
UnverifiedQwen3.5's parameter activation rate is 4.3%, while DeepSeek V3's 671B model has an activation rate of 5.5%.71% similarVerifiedQ4量化将每个权重参数从16位压缩到4位，理论上可将模型体积缩小约75%69% similarUnverifiedIn actual testing using the Qwen3 27B Q4 quantized dense model, the generation speed with thinking mode disabled is approximately 58–60 tokens/s.68% similarUnverifiedIn speculative decoding scenarios, the Q3 quantized version is actually slower than the Q4 quantized version.66% similarUnverifiedThe original Qwen3.6 27B Q4 scored 97 in the tool-calling test, demonstrating that distillation does not always improve performance.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/486835API
curl https://kongchang.com/api/v1/knowledge/claims/486835MCP
get_claim(id=486835)