one matrix replaced the KV cache. (the technique is 100% open source) Kimi just
Source
akshay_pachaar
Author
akshay_pachaar
Date
Why it matters
Attention mechanisms that eliminate KV cache growth change the cost model for long-context inference.
one matrix replaced the KV cache.
(the technique is 100% open source)
Kimi just dropped K3, an open model at frontier scale.
it leans on a new mechanism called delta attention that does not keep a growing KV cache. that is how it holds a million tokens of context without the https://t.co/zEDCWlmi20 https://t.co/JCdqjCKcKU
Transcript
one matrix replaced the KV cache.
(the technique is 100% open source)
Kimi just dropped K3, an open model at frontier scale.
it leans on a new mechanism called delta attention that does not keep a growing KV cache. that is how it holds a million tokens of context without the https://t.co/zEDCWlmi20 https://t.co/JCdqjCKcKU
