Verified65% confidenceFactExact time
V4.1 Flash activates only 8 billion parameters during the input stage and 16 billion parameters during the output stage.
3
Sources
65%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe local version of MAI Code 1.1 Flash has a total parameter count of 137 billion, activating only 6.8 billion parameters per inference.79% similarUnverifiedWorldofAI measured V4.1 Flash 1017's generation speed at about 244 tokens per second, with early tests approaching 300 tokens per second.72% similarUnverifiedV4.1 Flash 1017's lowest reasoning tier outputs 790 tokens, a 60% reduction from nearly 2000 tokens in the old version, and it can still answer questions correctly.72% similarVerifiedThe approximately 20:1 ratio between Qwen3 Next Flash's 125 billion total parameters and 6 billion activated parameters represents a fairly aggressive sparsity design.72% similarUnverifiedFlash-series models typically adopt a parameter scale of under 70B.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/908279API
curl https://kongchang.com/api/v1/knowledge/claims/908279MCP
get_claim(id=908279)