# Kinference

**Canonical:** https://apis.io/providers/kinference-ai/  
**Website:** https://kinference.ai/  
**APIs profiled:** 2

Kinference (pronounced "kay-inference") is an open-source inference control plane for Kubernetes and vLLM, built by Buoyant, Inc. (the company behind Linkerd). It turns any GPU-enabled Kubernetes cluster into a distributed inference platform with a traffic gateway that accepts OpenAI and Anthropic inference APIs, per-tenant token rate limits, policy-driven model routing, KV cache-aware load balancing, semantic classification, model aliases, and Prometheus/Grafana metrics, configured declaratively through Kubernetes CRDs and the kinfctl CLI. A commercial Kinference Enterprise edition is offered by Buoyant.

## Kin Score — 30.8 / 100 (thin)

Scored 2026-11-06 under rubric 0.27.0.

| Facet | Score |
|---|---|
| Discoverability | 57.1 |
| Contract Quality | 12.0 |
| Contract Governance | 18.2 |
| Operational Transparency | 21.1 |
| Developer Ergonomics | 64.3 |
| Access Clarity | 21.1 |

Regulatory layer — **Horizontal (data, software, accessibility, platform)**: 15.5 (matched via fallback).

## Agent readiness — 7.9 (agent-aware)

| Dimension | Value |
|---|---|
| Spec Presence | no |
| Agentic Access | no |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | bearer |
| Idempotency | no |
| Error Semantics | documented |
| OpenAPI Examples | no |
| Rate Limit Signal | documented |
| Event Surface Described | no |
| Agent Skills | no |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |
| Delegated Identity | no |
| Protected Resource Metadata | no |
| Dynamic Client Registration | no |
| Agentic Commerce | no |
| Pagination Ratio | na |
| Operation Distinctiveness | na |
| MCP Tool Quality | na |

## Access

Enterprise — onboarding: unknown, pricing: enterprise, trial: no (confidence: high).

## APIs (2)

- **Kinference Picker gRPC API** — The kinference.v1 Picker gRPC service used by the Kinference router - given an inference request it picks a model and destination (a local inference engine pod or an off-cluster...
- **Kinference Kubernetes CRD API** — Kubernetes custom resources in the kinference.buoyant.io/v1alpha1 group (InferencePolicy, InferencePool, InferenceProvider, InferenceWorkerSet, plus GatewayBackend) used to decl...

## Security (2)

- **Kinference Ai Authentication** — 2 schemes
- **Kinference Ai Domain Security** — TLSv1.3

## Plans (1)

- **Kinference Ai Plans Pricing**

## Tags

Company, Inference, Kubernetes, vLLM, LLM Gateway, Rate Limiting, Load Balancing, Open Source, Platform Engineering

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/kinference-ai/). Scores are computed from the provider's own public artifacts under a published rubric.