09:48
2026-08-13
discuss.huggingface.co
artificial-intelligence
[Research/Code] Verified 384x KV-Cache Compression with Linear Complexity using Finite Scalar Quantization (UL-SMF)
A developer has open-sourced the Unified Latent-State Memory Fabric (UL-SMF), a hardware-software co-designed architecture that compresses standard FP32 KV caches into a 16-dimensional discrete latentβ¦