v0.7.0

✈️ Highlights

• Direct vLLM provider — deploy vLLM OpenAI-compatible servers natively, no upstream operator required (#265)
• Generic provider discovery APIs — /providers endpoints + capability metadata so providers are self-describing; foundation for fully dynamic provider UX (#283)
• llm-d gateway capabilities — provider-specific EPP image + scheduler config while the controller manages the InferencePool (#288)
• GPU inference throughput estimation — surface estimated tokens/sec for models on cluster GPUs (#311)
• Documentation website — Docusaurus site now published from /docs (#310)
• Simplified setup — setup-kaito + provider-agnostic setup-gateway Makefile targets (#306, #305)
• CPU-only Dynamo mocker — GPU-less E2E coverage in CI (#307)
• Status-message and KAITO detection fixes, ESLint flat configs, throughput tooling, and dependency bumps

What's Changed

• feat(provider): kaito provider rbac permissions by @pauldotyu in #304
• deps: bump the dev-dependencies group across 1 directory with 19 updates by @dependabot[bot] in #291
• feat(kaito): add setup-kaito target to install KAITO operator by @surajssd in #306
• feat(gateway): add provider-agnostic setup-gateway target by @surajssd in #305
• feat(dynamo): add CPU-only mocker mode for GPU-less E2E coverage by @surajssd in #307
• ci: bump docker/setup-qemu-action from 4.0.0 to 4.1.0 by @dependabot[bot] in #313
• ci: bump actions/checkout from 6.0.2 to 6.0.3 by @dependabot[bot] in #312
• deps: bump the production-dependencies group with 8 updates by @dependabot[bot] in #315
• feat(website): docusaurus site published from /docs by @robert-cronin in #310
• deps: bump the dev-dependencies group across 1 directory with 4 updates by @dependabot[bot] in #316
• fix(backend): honour kubeconfig CA under Bun's native fetch by @surajssd in #319
• chore(scripts): make test-kv-routing.sh generic for any OpenAI endpoint by @ritazh in #263
• ci: bump actions/checkout from 6.0.2 to 6.0.3 by @dependabot[bot] in #321
• feat: estimate inference throughput for models on cluster GPUs by @surajssd in #311
• fix(kaito): detect KAITO installed via AKS AI-toolchain add-on by @surajssd in #322
• build(lint): add ESLint flat configs for both workspaces by @surajssd in #320
• feat: add direct vLLM provider support by @sozercan in #265
• fix(providers): refresh status message when ModelDeployment reaches Running by @robert-cronin in #326
• ci: bump actions/cache from 5.0.5 to 6.0.0 by @dependabot[bot] in #328
• ci: bump actions/checkout from 6.0.3 to 7.0.0 by @dependabot[bot] in #327
• feat: add generic provider discovery APIs by @sozercan in #283
• feat: Define provider specific gateway capabilities for llm-d by @ericdbishop in #288

Full Changelog: v0.6.0...v0.7.0

v0.6.0

What's Changed

• ci: bump docker/login-action from 4.0.0 to 4.1.0 by @dependabot[bot] in #205
• build(deps): bump go.opentelemetry.io/otel/sdk from 1.40.0 to 1.43.0 in /providers/kaito by @dependabot[bot] in #209
• build(deps): bump go.opentelemetry.io/otel/sdk from 1.40.0 to 1.43.0 in /controller by @dependabot[bot] in #210
• build(deps): bump go.opentelemetry.io/otel/sdk from 1.40.0 to 1.43.0 in /providers/dynamo by @dependabot[bot] in #211
• build(deps): bump go.opentelemetry.io/otel/sdk from 1.39.0 to 1.43.0 in /providers/kuberay by @dependabot[bot] in #212
• fix: update Hugging Face avatar URL to use absolute path by @pauldotyu in #220
• docs(azure): remove outdated GPU VM sizes by @pauldotyu in #227
• fix: Remove volume relying on azurefile-premium storage class in dynamo e2e test data by @ericdbishop in #207
• fix: update KAITO version from 0.9.0 to 0.10.0 in Helm config by @pauldotyu in #228
• docs(amlfs): fix query parameters for AMLFS client info by @pauldotyu in #234
• fix(dynamo): prevent panic on nodeSelector deep copy by @raphaelluchini in #231
• feat: delegate InferencePool + EPP management to providers with native gateway support by @ritazh in #213
• Add Playwright e2e tests for frontend UI by @robert-cronin in #186
• build(deps): bump go.opentelemetry.io/otel/sdk from 1.40.0 to 1.43.0 in /providers/llmd by @dependabot[bot] in #236
• fix: prevent provider crash-loop when backend CRDs are not installed by @robert-cronin in #224
• Add CODEOWNERS for airunway maintainers by @sozercan in #241
• ci: pin all actions and remove VEX generator workflows by @sozercan in #242
• ci: bump actions/upload-artifact from 4.6.2 to 7.0.1 by @dependabot[bot] in #222
• feat(dynamo): add setup-dynamo and cleanup-dynamo Makefile targets by @surajssd in #240
• ci: bump actions/cache from 4.2.3 to 5.0.5 by @dependabot[bot] in #248
• ci: bump docker/build-push-action from 7.0.0 to 7.1.0 by @dependabot[bot] in #249
• feat: move installation and documentation fields from InferenceProviderConfig spec to annotations by @vivekk16 in #180
• fix: update KubeRay default image to ray-llm:2.55.0-py311-cu128 by @robert-cronin in #237
• Add CNCF Community Code of Conduct v1.3 by @squillace in #256
• fix: update command line arguments in AI Configurator by @pauldotyu in #258
• adding governance file by @squillace in #255
• adding Apache 2 license with Microsoft signature to be removed later by @squillace in #253
• Fix runtime install status and provider CRD handling by @sozercan in #175
• Fix installed Inference Extension version display by @sozercan in #260
• fix(backend): require Gateway resource for gateway availability (#235) by @robert-cronin in #259
• fix(dashboard): update engine display logic by @pauldotyu in #266
• chore: add namespace to provider deployments by @pauldotyu in #257
• fix: resolve deployment metrics via frontend service by @sozercan in #261
• fix(kaito): finalize ModelDeployment when upstream CRD is missing by @robert-cronin in #267
• feat(ui): add gateway routing toggle to deployment form by @robert-cronin in #252
• Enable dashboard authentication by default in Kubernetes deployment by @sozercan in #273
• Harden llm-d override handling to block pod-spec privilege escalation by @sozercan in #272
• refactor(rbac): tighten dashboard ClusterRole to match exact code usage by @aramase in #191
• Fix deployment log pod selection by @sozercan in #270
• fix: support providers without runtime CRDs by @sozercan in #271
• fix: filter existing PVCs by deployment namespace by @robert-cronin in #225
• fix(ui): prevent manifest copy from submitting deploy form by @robert-cronin in #277
• chore(dynamo): bump Dynamo to 1.1.1 and install GAIE v1.5.0 by @surajssd in #280
• adding security draft by @squillace in #279
• feat(headlamp): upgrade plugin SDK and add gateway, disaggregated serving, and deployment enhancements by @gambtho in #184
• security: harden API, webhook, CI, and agent instructions by @robert-cronin in #206
• fix(kaito): deploy models that require GPU vs CPU by @ritazh in #262
• refactor: Centralize upstream versions by @surajssd in #281
• Add gambtho as Headlamp plugin code owner by @Copilot in #286
• fix(kaito): detect Eno-partial installs and refuse-fast on missing upstream controller by @gambtho in #217
• feat: add deployment chat panel by @sozercan in #275
• feat(api)!: restructure ProviderCapabilities to per-engine capability mapping by @surajssd in #214
• ci: bump docker/login-action from 4.1.0 to 4.2.0 by @dependabot[bot] in #299
• ci: bump docker/metadata-action from 6.0.0 to 6.1.0 by @dependabot[bot] in #298
• ci: bump docker/build-push-action from 7.1.0 to 7.2.0 by @dependabot[bot] in #296
• ci: bump docker/setup-buildx-action from 4.0.0 to 4.1.0 by @dependabot[bot] in #297
• deps: bump the production-dependencies group across 1 directory with 8 updates by @dependabot[bot] in #278
• feat(o11y): implement prometheus metrics by @pauldotyu in #284
• fix(prometheus): update ServiceMonitor configuration by @pauldotyu in #301
• fix(release): regressions in dockerfiles by @robert-cronin in #302

New Contributors

• @pauldotyu made their first contribution in #220
• @ericdbishop made their first contribution in #207
• @raphaelluchini made their first contribution in #231
• @robert-cronin made their first contribution in #186
• @squillace made their first contribution in #256
• @aramase made their first contribution in #191
• @Copilot made their first contribution in #286

Full Changelog: v0.5.0...v0.6.0

v0.5.0

What's Changed

• build(deps): bump google.golang.org/grpc from 1.78.0 to 1.79.3 in /providers/dynamo by @dependabot[bot] in #168
• ci: fix Dependabot automerge conditions by @sozercan in #172
• build(deps): bump google.golang.org/grpc from 1.78.0 to 1.79.3 in /providers/llmd by @dependabot[bot] in #169
• build(deps): bump google.golang.org/grpc from 1.78.0 to 1.79.3 in /controller by @dependabot[bot] in #171
• ci: retry Dependabot automerge on transient failures by @sozercan in #173
• Add demo to readme by @ritazh in #176
• feat: Add skill for /deploy-controller by @ritazh in #185
• ci: bump actions/setup-go from 6.3.0 to 6.4.0 by @dependabot[bot] in #188
• build(deps): bump google.golang.org/grpc from 1.78.0 to 1.79.3 in /providers/kaito by @dependabot[bot] in #189
• fix: HF token, model deployment mem limit, truncate long mode… by @ritazh in #182
• Filter gated HuggingFace models from search results when not logged in by @ritazh in #190
• Add VEX generator agentic workflow by @sozercan in #192
• ensure gateway allows modedeployment namespaces by @ritazh in #153
• fix: add namespace-level RBAC filtering for deployment listing (#99) by @vivekk16 in #130
• test: add backend API end-to-end tests by @gambtho in #196
• Update Dynamo provider for v1.0.0 by @sozercan in #148

New Contributors

• @vivekk16 made their first contribution in #130
• @gambtho made their first contribution in #196

Full Changelog: v0.4.0...v0.5.0

v0.4.0

What's Changed

• ci: bump actions/checkout from 6.0.1 to 6.0.2 by @dependabot[bot] in #112
• ci: bump docker/build-push-action from 6.18.0 to 6.19.2 by @dependabot[bot] in #113
• refactor: move deploy manifests and provider targets to per-provider structure by @sozercan in #111
• Add Gateway BBR demo with two models and BYO HTTPRoutes by @keithmattix in #102
• docs: clarify provider shim installation requirement by @sozercan in #117
• feat(ui): gateway endpoint display and CRD installation UI by @sozercan in #114
• fix: make HuggingFace token optional for non-gated models by @sozercan in #119
• ci(model-downloader): add Docker image build and release pipeline by @surajssd in #126
• feat(frontend): Midnight Cyberpunk UI redesign by @sozercan in #132
• fix(frontend): apply missing UI redesign changes from #132 by @sozercan in #140
• feat(controller,dynamo): add persistent storage support with PVC lifecycle and model download by @surajssd in #124
• fix(frontend): GPU fit indicator accounts for multi-GPU nodes by @sozercan in #144
• feat(ui): add storage volumes support for Dynamo deployments by @sozercan in #142
• feat: auto-infer multi-node deployment from model size by @sozercan in #143
• feat: update KAITO provider to v0.9.0 by @sozercan in #120
• fix(controller): increase model download job resources by @sozercan in #145
• Fix gateway and router issues by @ritazh in #141
• rename to AIRunway by @ritazh in #147
• fix gw for dynamo by @ritazh in #149
• Infer Dynamo vLLM pipeline parallelism for multi-node deployments by @sozercan in #150
• build: add optional push and platform flags to image targets by @sozercan in #152
• update AI Runway everywhere by @ritazh in #154
• ci: bump docker/metadata-action from 5.10.0 to 6.0.0 by @dependabot[bot] in #135
• ci: bump docker/build-push-action from 6.19.2 to 7.0.0 by @dependabot[bot] in #134
• ci: bump docker/setup-qemu-action from 3.7.0 to 4.0.0 by @dependabot[bot] in #136
• fix(dynamo): set vLLM side channel host from pod IP by @sozercan in #151
• Remove redundant refresh buttons and fix HuggingFace login prompt flash by @sozercan in #155
• Remove hardcoded provider colors from Settings runtime cards by @sozercan in #157
• fix(ui): use full cluster GPU capacity in fit badges by @sozercan in #158
• ci: bump oven-sh/setup-bun from 2.1.2 to 2.2.0 by @dependabot[bot] in #156
• ci: bump docker/setup-buildx-action from 3.12.0 to 4.0.0 by @dependabot[bot] in #137
• ci: bump docker/login-action from 3.7.0 to 4.0.0 by @dependabot[bot] in #138
• Fix Hugging Face settings UI readability by @sozercan in #161
• Scrub runtime references from frontend tests and mocks by @sozercan in #160
• Reduce /metrics request pressure on inference services by @sozercan in #162
• Fix deployment port-forward commands for runtime service ports by @sozercan in #164
• Improve KAITO deployment normalization and gateway guidance by @sozercan in #165
• Fix KAITO CPU manifest preview showing GPU resources by @sozercan in #166
• Normalize engine args and fallback deployment status by @sozercan in #170
• ci: add Dependabot automerge workflow by @sozercan in #159

New Contributors

• @surajssd made their first contribution in #126

Full Changelog: v0.3.1...v0.4.0

v0.3.1

What's Changed

• feat: upgrade to kaito 0.8.0 by @sozercan in #44
• Add manifest preview to model and deployment page by @Jont828 in #42
• feat: add prompt request workflow for AI-assisted contributions by @sozercan in #48
• Add realtime cost estimation for nodepools using Azure public pricing API by @Jont828 in #47
• fix: dynamo GPU configuration handling by @sozercan in #53
• Headlamp jonathan by @Jont828 in #55
• Update README to rename to KubeAIRunway by @ritazh in #63
• Add InferenceSet deployment option for KAITO by @keithmattix in #50
• docs: crd design by @sozercan in #58
• feat: provider implementation by @sozercan in #64
• feat(ui): replace Kubernetes-specific terminology with platform-agnostic language by @sozercan in #70
• Add architecture diagram and script by @ritazh in #72
• Update logo, add favicon, fix build, and add CI compile check by @sozercan in #74
• Add controller e2e integration tests with CPU-only inference by @sozercan in #75
• ci: bump docker/setup-buildx-action from 3.10.0 to 3.12.0 by @dependabot[bot] in #22
• ci: bump docker/login-action from 3.4.0 to 3.7.0 by @dependabot[bot] in #60
• ci: bump docker/setup-qemu-action from 3.6.0 to 3.7.0 by @dependabot[bot] in #17
• ci: bump docker/metadata-action from 5.7.0 to 5.10.0 by @dependabot[bot] in #14
• docs: add development status notice to README by @sozercan in #76
• ci: bump oven-sh/setup-bun from 2.0.2 to 2.1.2 by @dependabot[bot] in #61
• feat: integrate Gateway API Inference Extension for unified inference routing by @sozercan in #73
• fix: hide runtime installation details when no providers are registered by @sozercan in #100
• Fix build and return all ns by @ritazh in #90
• Separate controller and UI dashboard deployment manifests by @sozercan in #93
• ci: bump actions/setup-go from 5.5.0 to 6.2.0 by @dependabot[bot] in #86
• ci: bump actions/checkout from 6.0.1 to 6.0.2 by @dependabot[bot] in #88
• ci: bump docker/build-push-action from 6.18.0 to 6.19.2 by @dependabot[bot] in #87
• Bump go.opentelemetry.io/otel/sdk from 1.39.0 to 1.40.0 in /controller by @dependabot[bot] in #98
• Add llm-d provider by @ritazh in #92
• Bump go.opentelemetry.io/otel/sdk from 1.39.0 to 1.40.0 in /providers/llmd by @dependabot[bot] in #106
• fix: remove hardcoded provider count assertion in runtimes status test by @sozercan in #107
• feat: add e2e backend integration test workflow by @sozercan in #108
• Add backend API end-to-end tests by @sozercan in #109
• fix: add headlamp workspace to Dockerfile for bun install by @sozercan in #110

New Contributors

• @Jont828 made their first contribution in #42
• @keithmattix made their first contribution in #50

Full Changelog: v0.2.0...v0.3.1

v0.2.0

What's Changed

• Remove duplicating manual install section by @ritazh in #32
• Rename Ray Serve and rm installationPage by @ritazh in #35
• expand kaito support by @sozercan in #36
• test: add comprehensive unit tests for services and lib modules by @sozercan in #37
• feat: aiconfigurator by @sozercan in #38

New Contributors

• @ritazh made their first contribution in #32

Full Changelog: v0.1.0...v0.2.0

v0.1.0

What's Changed

• ci: bump actions/checkout from 4.3.1 to 6.0.1 by @dependabot[bot] in #1
• refactor by @sozercan in #4
• autoscaler by @sozercan in #13
• unified providers by @sozercan in #19
• add kaito by @sozercan in #21

New Contributors

• @dependabot[bot] made their first contribution in #1
• @sozercan made their first contribution in #4

Full Changelog: https://github.com/sozercan/kube-foundry/commits/v0.1.0