Expired50% confidenceBenchmarkExact time
On a task with one million input tokens, thirty thousand output tokens, and a 99% cache hit rate, MiMo-V2.6 Flash costs less than half of Qwen 3.8 Flash.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/23/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/7/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedCompared to 3.5 Flash, 3.6 Flash reduces average output tokens by approximately 17%, with some code tasks seeing reductions of up to 65%.77% similarUnverifiedMiMo V2.6 Flash has 310B total parameters and 15B activated parameters, with a speed nearly twice that of Pro.74% similarUnverifiedV4.1 Flash activates only 8 billion parameters during the input stage and 16 billion parameters during the output stage.72% similarUnverifiedThe DeepSeek Flash model can output more than 100 tokens per second.72% similarUnverifiedOn a single-GPU machine with 96 GB of VRAM and 188 GB of system memory, Qwen3.8 Flash Next was successfully run by offloading the embedding table and Ngram to system memory.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/944071API
curl https://kongchang.com/api/v1/knowledge/claims/944071MCP
get_claim(id=944071)