# ovg-project/kvcached

Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond

Repository: https://github.com/ovg-project/kvcached
Canonical: https://ross.abutalabs.com/products/kvcached
Language: Python
License: Apache-2.0
License Family: permissive
Topics: gpu-sharing, kvcache, llm, sglang, vllm, kvcached, inference-engine, gpu-mutiplexing, kvcache-optimization, llm-framework, llm-inference, llm-serving, serverless, ollama, elastic-kvcache, online-offline-coserve
Last push: 2026-08-23T05:24:49+00:00

## Health v2 (maintenance only)
Score: 74/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 66, longevity 33
- inputs: {"age_days": 463, "days_push": 10, "days_rel": 148, "gap_med": 43.0, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1143, forks 133 (observed 2026-08-28T04:03:45.028912+00:00)

## What it is
kvcached is a Python library that brings OS-style virtual memory abstraction to KV cache management for LLM serving and training on shared GPUs. It decouples GPU virtual addressing from physical memory allocation, enabling elastic, demand-driven KV cache allocation and better GPU utilization under dynamic workloads.

## Use cases
- share one GPU across multiple LLM serving instances
- reduce GPU memory waste from pre-allocated KV caches
- run vLLM or SGLang with elastic KV cache allocation
- co-serve online and offline LLM workloads on the same GPU
- improve GPU utilization under dynamic LLM traffic
- on-demand physical GPU memory allocation for KV caches

## When to choose
- you serve multiple LLM instances on shared GPUs and need flexible memory allocation
- you use vLLM or SGLang and want elastic, demand-driven KV cache management
- you need to mix online and offline LLM workloads on the same hardware
- GPU memory utilization is a cost concern in your LLM serving stack

## When to avoid
- you run a single LLM instance with plenty of dedicated GPU memory
- your inference engine is not vLLM or SGLang compatible
- you need a turnkey serving solution rather than a cache management layer
- you work on non-CUDA hardware without GPU virtual memory support

## Facets
- artifact type: library
- maturity: active
- function: llm-inference, caching, gpu-computing, machine-learning
- domain: large-language-models, gpu-computing, machine-learning, performance
- platform: python
- tags: kv-cache, gpu-sharing, vllm, sglang, virtual-memory, memory-management, llm-serving, inference-engine, gpu, linux, cuda

## Member repositories
- ovg-project/kvcached (main) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:45.028912+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:34:56.861900+00:00, confidence not recorded.
  - readme: https://github.com/ovg-project/kvcached (fetched 2026-08-28T04:03:45.028912+00:00, sha c7cc07696fbf)
  - registry_pypi: https://pypi.org/pypi/kvcached/json (fetched 2026-08-29T12:40:06.729649+00:00, sha b3d7fe3a5457)
- Data as of 2026-08-30T08:39:29.467469+00:00.