Request to Add CacheCraft: A Relevant Work on Chunk-Aware KV Cache Reuse

Thanks for this great list! We’d love to add CacheCraft [PDF]—a chunk-aware KV reuse approach for RAG that minimizes redundant computation while preserving generation quality. Our work is concurrent to CacheBlend, with key differences in chunk-level reuse, selective recompute planning, and optimizations designed for real-world production systems. CacheCraft is accepted at SIGMOD 2025. We’re also open-sourcing a vLLM-based extension soon. Results on real RAG traces show strong efficiency gains in production.

add the LLaMA-Factory project

add deepseek technical reports

Patch 1

Add paper: DeFT: Flash Tree-attention with IO-Awareness for Efficient Tree-search-based LLM Inference

Add mlc-llm

Add mobile papers

Add PowerInfer-2 and mlln-NPU

add PatrickStar

Add SGLang

Merge the distributed/single-node serving paper list since nowadays almost all new systems support multi-GPU processing with tensor or pipeline parallelism. Also added a new paper from Cal.