TurboQuant: Near-optimal KV cache quantization for LLM inference (3-bit keys, 2-bit values) with Triton kernels + vLLM integration
Like 0xSero/turboquant, 0 likes 0 Bookmark 0xSero/turboquant, 0 bookmarks 0 [Neurips 2025] R-KV: Redundancy-aware KV Cache Compression for Reasoning Models
Like Zefan-Cai/R-KV, 0 likes 0 Bookmark Zefan-Cai/R-KV, 0 bookmarks 0 SpectralQuant: Calibrated Eigenbasis Rotation and Water-Filled Bit Allocation for KV-Cache Compression
Like Dynamis-Labs/spectralquant, 0 likes 0 Bookmark Dynamis-Labs/spectralquant, 0 bookmarks 0 Unified KV Cache Compression Methods for Auto-Regressive Models
Like Zefan-Cai/KVCache-Factory, 0 likes 0 Bookmark Zefan-Cai/KVCache-Factory, 0 bookmarks 0 LLM KV cache compression made easy
Like NVIDIA/kvpress, 0 likes 0 Bookmark NVIDIA/kvpress, 0 bookmarks 0 LMCache: Supercharge Your LLM with the Fastest KV Cache Layer
Like LMCache/LMCache, 0 likes 0 Bookmark LMCache/LMCache, 0 bookmarks 0 Differential-KV (DKV) is a sparse KV-cache inference runtime designed for high-efficiency, memory-bounded long-context Large Language Model (LLM) inference across Apple Silicon (MLX) and CUDA GPUs.
Like Omc12/Differential-KV, 0 likes 0 Bookmark Omc12/Differential-KV, 0 bookmarks 0 FlashMemory DS-V4 Retriever: a lightweight retriever that sparsifies DeepSeek-V4 CSA KV-cache. Weights available on Hugging Face.
Like libertywing/FlashMemory-Deepseek-V4, 0 likes 0 Bookmark libertywing/FlashMemory-Deepseek-V4, 0 bookmarks 0 
