arXiv:2608.30963v1 Announce Type: cross Abstract: Modern large language model (LLM) serving systems increasingly operate over repeated or shared context, yet each model typically performs its own pre...
arXiv:2608. 01247v1 Announce Type: cross Abstract: Query-agnostic KV cache eviction compresses a context once and reuses the resulting cache for arbitrary future queries, but performance can collapse under tight budgets.
Query-agnostic KV cache eviction compresses a context once and reuses the resulting cache for arbitrary future queries, but performance can collapse under tight budgets. Existing methods primarily improve which original KV pairs are retained.
arXiv:2608. 11231v1 Announce Type: new Abstract: LLM serving is increasingly accelerated by position-independent caching (PIC).
The paper investigates whether training Mixture-of-Experts (MoE) routers can improve memory‑bandwidth locality on consumer GPUs. Using a new zero‑surgery telemetry tool, the authors measure that a large Qwen3‑235B model is bottlenecked by disk‑based expert access, and that an LRU cache can serve a majority of requests. They pre‑register experiments training 137 M‑parameter MoE models with locality‑aware losses, finding that while cache misses can drop up to 60 % (99 % static‑pin hit rate), every configuration fails to meet a strict 1 % perplexity threshold, indicating a tight coupling between cache efficiency and model quality.
arXiv:2609.10266v1 Announce Type: new Abstract: LLM serving systems already reuse KV caches, but only when the reused text sits at the very start of the prompt. Two growing workloads break this condi...
Long-context inference retains a growing key--value (KV) cache during decoding, which consumes substantial GPU memory and can reduce generation throughput. This bottleneck remains in hybrid language m...
The paper investigates how to recover language model quality lost when using low‑bit key–value caches for autoregressive decoding. By keeping the quantizer fixed and distilling the full‑precision cache behavior into low‑rank Q/K/V projection updates, the authors demonstrate that 4‑bit affine‑cache adapters recover roughly 54 % of the perplexity gap on TinyLlama‑1.1B and 76 % on Gemma‑4‑12B, while preserving most long‑context retrieval. Even a 2‑bit rank–token sweep can dramatically reduce TinyLlama’s perplexity, though it only partially restores retrieval performance.
HeadWiseKV is a training‑free framework that compresses the residual global key–value caches of hybrid long‑context language models by assigning each physical KV head a static, multilevel history window. It formulates cache allocation as a restricted operational rate–distortion problem and uses the SeqCalib algorithm to generate per‑head residency policies that account for interactions across layers. In evaluations on four hybrid models, HeadWiseKV preserves near‑full‑KV quality while reducing peak device memory usage by 8.59% at a 112K context length and extending the largest verified context from 114K to 161K.
arXiv:2609.14507v1 Announce Type: cross Abstract: Single-GPU long-context inference with Mixture-of-Experts (MoE) models requires spilling the key-value cache (KVCache) to CPU memory. The spilled KV...
The paper presents py‑kvcache, a new KV offload connector for vLLM that uses asynchronous direct I/O, bounded shared staging, and scheduler‑aware preloading to improve external KV caching performance on NVMe SSDs. Experiments across synthetic workloads, long‑context benchmarks, and production traces show that py‑kvcache can load 80k‑token prefixes 2.0× faster than LMCache, with preloading contributing an additional 1.34× speedup, and achieves overall performance within 4% of native vLLM KV Offload. The study highlights that cache effectiveness depends on transfer granularity, intermediate memory use, and scheduling timing rather than just device bandwidth, indicating that external KV caching should be considered a setup‑specific admission decision.
Serving a 235B-parameter Mixture-of-Experts (MoE) model on a single 8 GB GPU is bottlenecked not by compute but by memory bandwidth: decode must stream each token's active experts from whichever tier holds them, and on consumer hardware most experts sit on an SSD far slower than RAM. We quantify this bandwidth wall on Qwen3-235B (Q4_K_M, 134 GB): measured decode is 0.