# llm-d/llm-d

Achieve state of the art inference performance with modern accelerators on Kubernetes

Repository: https://github.com/llm-d/llm-d
Canonical: https://ross.abutalabs.com/products/llm-d
Homepage: https://www.llm-d.ai
Language: Shell
License: Apache-2.0
License Family: permissive
Topics: ai, cncf, distributed-inference, gpu, inference, intelligent-routing, kubernetes, llm, model-server
Last push: 2026-08-26T22:59:24+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 86, longevity 35
- inputs: {"age_days": 491, "days_push": 7, "days_rel": 16, "gap_med": 34.0, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4164, forks 706 (observed 2026-08-28T04:08:37.694757+00:00)

## What it is
llm-d is a Kubernetes-native distributed LLM inference serving stack that orchestrates model servers like vLLM and SGLang across clusters. It provides intelligent prefix-cache and load-aware routing, tiered KV-cache management, prefill/decode disaggregation, wide expert parallelism, and SLO-aware autoscaling, packaged as tested 'Well-Lit Path' Helm deployment recipes.

## Use cases
- serve llm inference at scale on kubernetes
- route requests by prefix cache affinity
- deploy deepseek-r1 with wide expert parallelism
- autoscale llm inference pool based on slo
- offload kv cache to cpu and disk
- run vllm across multiple gpus in production
- reduce time to first token for multi-turn chat
- batch offline inference with openai-compatible api

## When to choose
- you already run Kubernetes and want production-grade distributed LLM serving on GPUs, TPUs, or other accelerators
- you need LLM-aware load balancing, KV-cache reuse, or disaggregated prefill/decode beyond what a single vLLM/SGLang node provides
- you want vendor-neutral, engine-agnostic orchestration with tested deployment recipes and benchmarks

## When to avoid
- you serve a single model on one node and vanilla vLLM or SGLang meets your needs
- you have no Kubernetes infrastructure and don't want to adopt it
- you need a simple managed API endpoint rather than operating your own inference cluster

## Facets
- artifact type: framework
- maturity: active
- function: llm-inference, api-gateway, load-testing, caching, monitoring, deployment, container-orchestration
- domain: large-language-models, machine-learning, gpu-computing, cloud-computing, microservices, infrastructure-as-code
- platform: cloud, self-hosted, go, python
- tags: distributed-inference, kv-cache, vllm, sglang, intelligent-routing, prefix-cache, moe, cncf-sandbox, helm-charts, autoscaling, gateway-api-inference-extension, containers, kubernetes, docker, gpu

## Member repositories
- llm-d/llm-d (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:37.694757+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 2026-08-29T18:22:47.453709+00:00, confidence not recorded.
  - readme: https://github.com/llm-d/llm-d (fetched 2026-08-28T04:08:37.694757+00:00, sha b0c6d70bb3c3)
  - homepage: https://www.llm-d.ai (fetched 2026-08-29T09:13:42.091244+00:00, sha b60fc371f228)
  - site_page: https://llm-d.ai/docs/getting-started/quickstart (fetched 2026-08-29T09:13:42.100078+00:00, sha 02d870a95c0f)
  - site_page: https://llm-d.ai/docs (fetched 2026-08-29T09:13:42.101955+00:00, sha 1e00c66d2397)
  - site_page: https://llm-d.ai/docs/0.8 (fetched 2026-08-29T09:13:42.103501+00:00, sha 1d40cbf5dfd6)
  - site_page: https://llm-d.ai/docs/0.7 (fetched 2026-08-29T09:13:42.105025+00:00, sha c81c2da56f78)
  - site_page: https://llm-d.ai/docs/dev (fetched 2026-08-29T09:13:42.106810+00:00, sha 54f751bb69db)
  - site_page: https://llm-d.ai/docs/well-lit-paths/foundations/optimized-baseline (fetched 2026-08-29T09:13:42.108359+00:00, sha 3a28a2cbb49b)
  - site_page: https://llm-d.ai/docs/well-lit-paths/foundations/predicted-latency (fetched 2026-08-29T09:13:42.110140+00:00, sha 9510eedda65a)
  - site_page: https://llm-d.ai/docs/well-lit-paths/foundations/precise-prefix-cache-routing (fetched 2026-08-29T09:13:42.111912+00:00, sha cd39e1aaff99)
- Data as of 2026-08-30T08:39:29.467469+00:00.