Unverified50% confidenceFactExact time
After quantizing Flash-Next's expert layers to 3-bit, the score dropped by only 0.02, and token consumption increased by only 3%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/2/2026
First Seen
Valid until: 1/31/2027
Sources
Related Entities
Related Claims
UnverifiedV4.1 Flash offers three levels of reasoning intensity—low, medium, and highest—with the average score across eight reasoning benchmarks rising from 67.1 to 76.3, at the cost of approximately 2.5 times more output tokens.70% similarUnverifiedThe activated parameter count of Qwen3.8 Flash is approximately one-third that of GLM.67% similarUnverified盘古2.0 Flash在AIME类数学测试中得分93.3，在HMMT测试中得分91.5，数学能力处于第一梯队66% similarPartially VerifiedV4.1 Flash 1017's score on Terminal-Bench 4.0 rose from 31.2% to 42.3%, a relative improvement of 35% in five weeks.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1127041API
curl https://kongchang.com/api/v1/knowledge/claims/1127041MCP
get_claim(id=1127041)