Introducing Kinference: An Open Source Inference Control Plane for vLLM and Kubernetes
First announcement: an open source control plane covering token rate limiting, tenant usage and cost tracking, and load balancing for vLLM.
AI Storyline
3 items · 3 sources · 3 days
Operational story trace
Inference Kubernetes
Follow in this browser to see new updates on your Live feed.
Current stateEmerging
Latest change
An InfoQ podcast (Nov 9) places llm-d and vLLM on Kubernetes in the wider distributed-inference conversation, alongside WebAssembly on the backend and TornadoVM for Java on GPUs.
Earlier contextThe story so far
On Nov 4 the vLLM project introduced Kinference, an open source inference control plane for vLLM on Kubernetes. A benchmark post comparing it against plain Kubernetes followed a day later.
editor-curated · source-linked
Arc
Nov 4Nov 9 · now
LAUNCH · Nov 4Kinference debuts as an inference control plane for vLLM on Kubernetes1 source · show source ▾
EVIDENCE · Nov 5Buoyant benchmarks Kinference against plain Kubernetes1 source · show source ▾
Distributed Inference Benchmarks: [K]Inference vs. Plain Kubernetes
A benchmark post comparing Kinference with plain Kubernetes, shared on Hacker News.
NOW · Nov 9Practitioner view: llm-d and vLLM on Kubernetes, plus Wasm and TornadoVM1 source · show source ▾
Podcast: Distributed LLM Inference on Kubernetes with llm-d, Wasm, & TornadoVM
Podcast with Edoardo Vacchi covering llm-d and vLLM on Kubernetes, widening the thread beyond Kinference.
Day 1 Wednesday, Nov 4, 2026
Introducing Kinference: An Open Source Inference Control Plane for vLLM and Kubernetes
First announcement: an open source control plane covering token rate limiting, tenant usage and cost tracking, and load balancing for vLLM.
Day 2 Thursday, Nov 5, 2026
Distributed Inference Benchmarks: [K]Inference vs. Plain Kubernetes
A benchmark post comparing Kinference with plain Kubernetes, shared on Hacker News.
Day 3 Monday, Nov 9, 2026
Podcast: Distributed LLM Inference on Kubernetes with llm-d, Wasm, & TornadoVM
Podcast with Edoardo Vacchi covering llm-d and vLLM on Kubernetes, widening the thread beyond Kinference.
What to watch — open questions
- Does an independent benchmark, not from the Kinference or Buoyant side, reproduce the comparison against plain Kubernetes?
- How do Kinference and llm-d compare on features such as load balancing and tenant usage tracking?
How this thread was built
editor wrote the arc · 3 beatswatcher 1 status change
Storylines are threaded mechanically from the feed: stories that share a distinctive anchor across multiple days and sources. Each item links to its original source. The evidence trace, current state, and open questions are written by the editor routine and refreshed whenever a new beat lands.