HAMi website screenshot

HAMi

HAMi (Heterogeneous AI Computing Virtualization Middleware) is a CNCF incubating open-source project that brings device sharing, memory and core isolation, and topology-aware scheduling to heterogeneous AI accelerators on Kubernetes. It lets several containers share one physical GPU or NPU with hard memory limits and enforced compute quotas, across NVIDIA, Huawei Ascend, Cambricon MLU, Hygon DCU, Enflame GCU, Iluvatar, Moore Threads, MetaX, Kunlunxin and Vastai devices. HAMi ships a scheduler, a device plugin, an in-container enforcement library (HAMi-core/libvgpu), a DRA driver, and HAMi WebUI — an observability dashboard whose backend exposes a gRPC and grpc-gateway REST API over cluster GPU, node, workload and Prometheus-backed monitoring data. The project also publishes a machine-readable discovery surface on its documentation site: an RFC 9727 API catalog, an OpenAPI 3.1 service description, an Agent Skills discovery index, and Content-Signal directives in robots.txt.

HAMi publishes 3 APIs on the APIs.io network, including Robots.txt API, .well Known API, and 1 more. Tagged areas include AI Computing, CNCF, GPU Virtualization, Kubernetes, and GPU Sharing.

HAMi’s developer surface includes documentation, engineering blog, getting-started guide, support, changelog, authentication, and 24 more developer resources.

🌐 Visit website 📡 Source on GitHub
40.8/100 developing ▬ flat Agent 31/100 agent ready Front door AI 3/6 light saas Full breakdown ↓
scored 2026-09-25 · rubric v0.23.0
1 APIs
AI ComputingCNCFGPU VirtualizationKubernetesGPU SharingSchedulingOpen SourceInfrastructureObservabilityHeterogeneous Computing

Kin Score

Kin Score Kin Score How this is scored →
scored 2026-09-25 · rubric v0.23.0
Composite quality — 40.8/100 · developing
Contract Quality 8.2 / 21
Developer Ergonomics 9.7 / 17
Access Clarity 3.6 / 17
Operational Transparency 3.8 / 11
Contract Governance 1.9 / 10
Discoverability 5.6 / 9
Regulatory Posture 4.6 / 15
Agent readiness — 31/100 · agent ready
Machine-Readable Contract 18 / 18
Agentic Access Contract 0 / 10
Documented Reversibility 6 / 6
MCP Server 0 / 12
Machine-Readable Auth 10 / 10
Idempotency 9 / 9
Stable Error Semantics 8 / 8
Request/Response Examples 0 / 7
Rate-Limit Signaling 7 / 7
Typed Event Surface 0 / 6
Agent Skills 5 / 5
Well-Known Catalog 4 / 4
Consent & Bot Identity 0 / 3
A2A Agent Card 0 / 8
Dry-Run / Simulate Mode 4 / 4
Delegated User Identity 0 / 6
Protected Resource Metadata 5 / 5
Registration Without a Human 0 / 6
Agentic Commerce Well-Known Document 0 / 5
Regulatory Posture applies to this provider. Its tags matched the Horizontal (data, software, accessibility, platform) regime, so Regulatory Posture carries 15 points of the composite. If this regime is wrong for your business, say so on your provider repo — the applicability map is public and we will correct it.
Create-or-Update Ergonomics does not apply to this provider. The published contracts declare no write operations, and a read-only API cannot create-or-update. The facet is excluded from this provider's denominator entirely — not scored zero. A reference or data API is not deficient for being unable to upsert.
The six quality facets above are damped to 85 points between them, because the conditional facet above carries the other 15. That is why each facet's contribution is shown against a damped maximum: raising a quality facet moves the composite by 85% of its nominal weight, not 100%. The full arithmetic is at apis.io/rating/.
Improve this rating by publishing the missing artifacts — every area above can be raised, and the full rubric is at apis.io/rating/. Every facet and dimension name above is a link: it opens that measurement's own page — what it means, the exact checks that feed it, how the whole catalog distributes on it, and the providers at the top of it. This rating is computed from github.com/api-evangelist/hami: open an issue to ask a question, or submit a pull request to add artifacts. Submit an artifact on GitHub — free → Manage your own listing — the Influence plan, $499/mo →

APIs 3

Individual APIs this provider publishes, each with its own machine-readable definition.

HAMi WebUI API

The HAMi WebUI backend API. Four gRPC services — Card, Node, Container and Monitor — expose cluster accelerator inventory, node inventory, GPU-consuming workloads and Prometheus...

HAMi Robots.txt API

The Robots.txt API from HAMi — 1 operation(s) for robots.txt.

HAMi .well Known API

The .well Known API from HAMi — 3 operation(s) for .well known.

Pricing Plans 1

Published pricing tiers and plan structures.

Hami Plans Pricing

0 plans

PLANS

Rate Limits 1

Documented rate limits and quota policies.

Hami Rate Limits

0 limits

RATE LIMITS

Security Posture 3

Authentication, domain security, vulnerability disclosure, and trust-center signals.

Hami Authentication

0 schemes

SECURITY

Hami Domain Security

TLSv1.3 · HSTS

SECURITY

Hami Vulnerability Disclosure

disclosure policy published

SECURITY

Resources

Get Started 2

Portal, sign-up, and the first successful call

DeveloperPortal
DeveloperPortal
 GettingStarted
GettingStarted

Documentation 1

Reference material describing how the API behaves

Documentation
Documentation

Agent Surfaces 4

MCP servers, agent skills, and machine-readable catalogs

WellKnown
WellKnown
 AgentSkill
AgentSkill
 LLMsTxt
LLMsTxt
 X-MCPServerCandidate
X-MCPServerCandidate

Design & Contract 5

Pagination, idempotency, versioning, errors, and events

Conformance
Conformance
 ErrorCatalog
ErrorCatalog
 Lifecycle
Lifecycle
 Conventions
Conventions
 DataModel
DataModel

Build 2

SDKs, sample code, and the tooling you integrate with

GitHubOrganization
GitHubOrganization
 Packages
Packages

Access & Security 5

Authentication, authorization, and security posture

Security
Security
 Authentication
Authentication
 DomainSecurity
DomainSecurity
 VulnerabilityDisclosure
VulnerabilityDisclosure
 X-SecurityInsights
X-SecurityInsights

Operate 5

Status, limits, changes, and where to get help

Support
Support
 Roadmap
Roadmap
 ChangeLog
ChangeLog
 ChangeLog
ChangeLog
 RateLimits
RateLimits

Commercial 3

Pricing, plans, and the legal terms of use

TermsOfService
TermsOfService
 PrivacyPolicy
PrivacyPolicy
 Plans
Plans

Company 2

The organization behind the API

Website
Website
 Blog
Blog

Other 1

Properties that don't map to a standard resource type

Overlay
Overlay

Source (apis.yml)

apis.yml Raw ↑ 
aid: hami
name: HAMi
description: 'HAMi (Heterogeneous AI Computing Virtualization Middleware) is a CNCF incubating open-source project that brings
  device sharing, memory and core isolation, and topology-aware scheduling to heterogeneous AI accelerators on Kubernetes.
  It lets several containers share one physical GPU or NPU with hard memory limits and enforced compute quotas, across NVIDIA,
  Huawei Ascend, Cambricon MLU, Hygon DCU, Enflame GCU, Iluvatar, Moore Threads, MetaX, Kunlunxin and Vastai devices. HAMi
  ships a scheduler, a device plugin, an in-container enforcement library (HAMi-core/libvgpu), a DRA driver, and HAMi WebUI
  — an observability dashboard whose backend exposes a gRPC and grpc-gateway REST API over cluster GPU, node, workload and
  Prometheus-backed monitoring data. The project also publishes a machine-readable discovery surface on its documentation
  site: an RFC 9727 API catalog, an OpenAPI 3.1 service description, an Agent Skills discovery index, and Content-Signal directives
  in robots.txt.'
url: https://raw.githubusercontent.com/api-evangelist/hami/refs/heads/main/apis.yml
type: Index
position: Producing
deliveryModel:
  model: saas
  license_evidence:
    not_product:
    - project-hami/hami-webui
  open_source: unknown
  commercial: true
  callable_host: true
  label: Hosted service · you call their endpoint
  confidence: low
  source:
  - openapi
  - pricing
  - repository-not-product
  generated: '2026-09-25'
  method: derived
accessModel:
  pricing: unknown
  onboarding: unknown
  trial: false
  try_now: false
  public: false
  label: Unknown
  confidence: low
  source: []
  generated: '2026-07-22'
  method: derived
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/icons/hami.png
tags:
- AI Computing
- CNCF
- GPU Virtualization
- Kubernetes
- GPU Sharing
- Scheduling
- Open Source
- Infrastructure
- Observability
- Heterogeneous Computing
tags_raw:
- AI Computing
- CNCF
- GPU Virtualization
- Kubernetes
- GPU Sharing
- Scheduling
- Open Source
- Infrastructure
- Observability
- Heterogeneous Computing
- Open-Source
created: '2025-01-01'
modified: '2026-09-12'
specificationVersion: '0.23'
apis:
- name: HAMi WebUI API
  description: The HAMi WebUI backend API. Four gRPC services — Card, Node, Container and Monitor — expose cluster accelerator
    inventory, node inventory, GPU-consuming workloads and Prometheus-backed range/instant/summary queries. Every RPC carries
    a google.api.http annotation, so the same surface is served as REST through grpc-gateway under /v1 (for example POST /v1/gpus,
    GET /v1/node, POST /v1/monitor/query/range-vector). The contract is the .proto set published in the HAMi-WebUI repository
    and mirrored verbatim into grpc/ here.
  humanURL: https://project-hami.io/docs/installation/webui-installation
  baseURL: http://{hami-webui-host}:8000
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/icons/hami.png
  tags:
  - GPU Virtualization
  - Kubernetes
  - Observability
  - Monitoring
  properties:
  - type: Protobuf
    url: grpc/hami-webui-card.proto
  - type: Protobuf
    url: grpc/hami-webui-node.proto
  - type: Protobuf
    url: grpc/hami-webui-container.proto
  - type: Protobuf
    url: grpc/hami-webui-monitor.proto
  - type: Protobuf
    url: grpc/hami-webui-error.proto
  - type: SourceCode
    url: https://github.com/Project-HAMi/HAMi-WebUI
  - type: Documentation
    url: https://project-hami.io/docs/installation/webui-installation
- aid: hami:hami-robots-txt-api
  name: HAMi Robots.txt API
  description: The Robots.txt API from HAMi — 1 operation(s) for robots.txt.
  humanURL: https://project-hami.io/docs/installation/webui-installation
  baseURL: http://{hami-webui-host}:8000
  tags:
  - Robots.txt
  properties:
  - type: OpenAPI
    url: openapi/hami-robots-txt-api-openapi.yml
  - type: WellKnown
    url: well-known/hami-well-known.yml
- aid: hami:hami-well-known-api
  name: HAMi .well Known API
  description: The .well Known API from HAMi — 3 operation(s) for .well known.
  humanURL: https://project-hami.io/docs/installation/webui-installation
  baseURL: http://{hami-webui-host}:8000
  tags:
  - .well Known
  properties:
  - type: OpenAPI
    url: openapi/hami-well-known-api-openapi.yml
  - type: WellKnown
    url: well-known/hami-well-known.yml
common:
- type: Overlay
  url: overlays/hami-website-discovery-overlay.yaml
- type: Website
  url: https://www.project-hami.io/
- type: GitHubOrganization
  url: https://github.com/Project-HAMi/HAMi
- type: Documentation
  url: https://project-hami.io/docs/
- url: https://project-hami.io/blog/rss.xml
  type: Blog
- type: DeveloperPortal
  url: https://project-hami.io/docs/category/developer-guide
- type: GettingStarted
  url: https://project-hami.io/docs/get-started/deploy-with-helm
- type: Support
  url: https://project-hami.io/community
- type: Roadmap
  url: https://project-hami.io/docs/contributor/roadmap
- type: ChangeLog
  url: https://project-hami.io/changelog
- type: TermsOfService
  url: https://lfprojects.org/policies/terms-of-use/
- type: PrivacyPolicy
  url: https://lfprojects.org/policies/privacy-policy/
- type: Security
  url: https://github.com/Project-HAMi/HAMi/blob/master/SECURITY.md
- type: WellKnown
  url: well-known/hami-well-known.yml
- type: AgentSkill
  url: skills/_index.yml
- type: LLMsTxt
  url: llms/hami-llms.txt
- type: Packages
  url: packages/hami-packages.yml
- type: Conformance
  url: conformance/hami-conformance.yml
- type: ErrorCatalog
  url: errors/hami-problem-types.yml
- type: Lifecycle
  url: lifecycle/hami-lifecycle.yml
- type: ChangeLog
  url: changelog/hami-changelog.yml
- type: Conventions
  url: conventions/hami-conventions.yml
- type: DataModel
  url: data-model/hami-data-model.yml
- type: Authentication
  url: authentication/hami-authentication.yml
- type: DomainSecurity
  url: security/hami-domain-security.yml
- type: VulnerabilityDisclosure
  url: security/hami-vulnerability-disclosure.yml
- type: X-SecurityInsights
  url: security/hami-security-insights.yml
- type: RateLimits
  url: rate-limits/hami-rate-limits.yml
- type: Plans
  url: plans/hami-plans-pricing.yml
- type: X-MCPServerCandidate
  url: mcp/hami-mcp.yml
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com
x-enrichment:
  date: '2026-09-12'
  status: enriched
  artifacts_added: 31
  pass: local-v3

Work with this as data

Every provider here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for providers

9 MCP tools reach this
  • find_providersBrowse and filter every provider in the catalog.
  • get_provider_artifactsEvery artifact this provider publishes, grouped by type.
  • get_provider_operationsEvery operation across all of their OpenAPIs — one call instead of parsing every spec.
  • get_provider_toolsEvery MCP tool they ship, with the operation each wraps.
  • get_provider_evidenceHow each part of their score was established. Free — the basis for a claim should not sit behind it.
  • get_provider_ratingPRO — composite, band, trend and facet scores.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools →

Call it yourself

curl for this page
This provider 
curl "https://apis.io/api/v1/providers/hami"
All providers 
curl "https://apis.io/api/v1/providers?limit=25"
Every operation they expose 
curl "https://apis.io/api/v1/providers/hami/operations?limit=25"
How their score was established 
curl "https://apis.io/api/v1/providers/hami/evidence"

Discovery needs no key. Ratings and market analysis are Pro.

Get an API key

Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.

GitHub Google LinkedIn

A second provider on the same verified email joins the account you already have.