Releases: Azure/kaito
Releases · Azure/kaito
v0.2.2
v0.2.2 - 2024-04-04
This release reverts a few default inference parameters back to the values used in v0.1.0 to avoid any user confusions. These parameters significantly impact the inference results.
Changelog
Bug Fixes 🐞
- 98bf904 fix: Update namespace in the helm chart (#337)
- 05b7a46 fix: Add registry as a job output (#328)
- e8ffbd1 fix: Update helm chart to use Release.Namespace (#322)
- 8af6207 fix: Fix typo in Makefile (#315)
- 7f4e683 fix: Update Model Tags (#311)
- f638608 fix: Adjust default model params (#310)
Continuous Integration 💜
Documentation 📘
v0.2.1
v0.2.1 - 2024-03-19
This release includes a critical fix to revert the default inference max sequence length back to 200 as it was in v0.1.0. A commit in v0.2.0 accidentally changes the default max sequence length to 20.
Changelog
Features 🌈
Bug Fixes 🐞
- ed345d6 fix: Protect secret with environment (#300)
- 4c4e803 fix: Update default params and add associated UTs (#294)
- 268675c fix: update manifest and helm charts (#278)
Continuous Integration 💜
- c704f84 ci: fix 1ES pool label name (#301)
- 63ff6cf ci: Update supported_models.yaml (#296)
- ab88635 ci: Add environment for pipelines (#290)
- dd59ef3 ci: Use 1ES runner for kaito workspace workflow jobs that push to ACR (#283)
Documentation 📘
- d58d22f docs: Add gpu-provisioner github repo (#267)
- 13c77e3 docs: update README.md for new models (#279)
Maintenance 🔧
- 9b33f33 chore: bump peter-evans/repository-dispatch from 1 to 3 (#269)
- c54a32b chore: bump azure/setup-helm from 3 to 4 (#270)
- 40a6e03 chore: bump actions/checkout from 3 to 4 (#271)
Security Fix 🛡️
- d5d3c57 Fix protobuf to address CVE-2024-24786
- ebf2f46 Fix fastapi to address CVE-2024-24762
v0.2.0
v0.2.0 - 2024-03-04
Changelog
Features 🌈
- 3d1dead feat: Add make target to run e2e local (#239)
- 42a5965 feat: Add FAQ Section to README (#263)
- 7bd5641 feat: Add additional e2e tests (#256)
- de2b45c feat: Add Image Pull Secret (#258)
- 37485fa feat: Add Mistral and Phi-2 model Plugins (#249)
- 68c972d feat: inference API cleanup (#233)
- 23c906f feat: Prevent unnecessary build re-triggers (#246)
- 133cab3 feat: Update Preset Configs (#243)
- 5e1ebb9 feat: phi-2 proposal (#235)
- cbdac27 feat: Mistral 7B Proposal (#231)
- 4a5dbef feat: Add Mistral 7b Instruct (#232)
- a70138e feat: Part 4 - Add image tag to models (#221)
- ae881d2 feat: Part 3 - Update E2E tests for new Build Preset Pipeline (#220)
- a9c766f feat: Part 2 - New build pipeline (#217)
- ecc56d1 feat: Organize Inference Files - Part 1 (#216)
- 7c88c4e feat: organize image info (#203)
- 95171e1 feat: ignore doc changes in build pipeline (#204)
- dc09bf1 feat: containerize build - part 2 (#188)
Bug Fixes 🐞
- c93868e fix: Add release history comments (#266)
- a299248 fix: Remove Llama Tag (#265)
- cc53389 fix: Switch to Production ACR (#252)
- 559efed fix: Eliminate Unnecessary Process Group Creation in Worker Initialization (#244)
- 2fb9387 fix: Update llama dockerfile (#248)
- 31aa7ac fix: Add Az Login
- 9e66cf8 fix: Minor build and e2e workflow fixes (#245)
- 83597c0 fix: Prevent cleanup function from killing all jobs
- 3bd8599 fix: Update main.py
- ba50963 fix: Part 5.5 - Add determine models code to e2e (#236)
- 7201161 fix: Reduce model image sizes (#225)
- fbdd2ec fix: Add resource requirements (#227)
- a2a039a fix: Minor Fix in e2e preset pipeline - Checking ACR Image (#223)
- 7856b76 fix: resolve git action paths (#205)
- b9abdb4 fix: Update kaito-e2e.yaml (#200)
- c7ff5b0 fix: remove extraneous params (#197)
- f93bc04 fix: install cli in e2e-test (#178)
Code Style 🎶
Continuous Integration 💜
- 04819b2 ci: Add scan trivy step to publish image pipeline (#257)
- 8c40ab0 ci: Part 4.5 - Fix Minor Bugs in E2E Preset Pipeline (#222)
- bdcac7a ci: trigger preset image build (#193)
- dcec033 ci: temp disable building new images (#186)
- 64ac7e7 ci: Support codcov (#172)
Documentation 📘
- 1e998e7 docs: Update installation.md (#255)
- 3b7eebe docs: move installation steps to a separate file. (#226)
- fe9a8ad docs: Add instructions about how to add new model in kaito (#224)
- 7e80975 docs: add new model proposal template (#211)
- f38c8b1 docs: update install instructions (#208)
- ba8d09a docs: correct wrong build commands (#206)
- 041ba52 docs: Revise preset model READMEs (#202)
- b9f01d9 docs: update README.md (#171)
Maintenance 🔧
- c4bc7aa chore: bump thehanimo/pr-title-checker from 1.4.1 to 1.4.2 (#240)
- d4e23a9 chore: bump fastapi from 0.103.2 to 0.109.1 in /presets/inference/text-generation (#230)
- df71240 chore: bump step-security/harden-runner from 2.6.1 to 2.7.0 (#229)
- 86d0f46 chore: Part 5 - documentation updates (#218)
- 2255416 chore: bump codecov/codecov-action from 3 to 4 (#228)
- 7918029 chore: bump azure/login from 1.6.0 to 1.6.1 (#214)
- c9a63bc chore: bump azure/login from 1.5.1 to 1.6.0 (#209)
- 3d1754f chore: bump transformers from 4.33.3 to 4.36.0 in /presets/models/falcon (#195)
- 39e9b4c chore: bump azure/login from 1.4.6 to 1.5.1 (#199)
- c5928aa chore: bump actions/setup-go from 4 to 5 (#182)
- 759b3a0 chore: bump actions/upload-artifact from 3 to 4 (#192)
- 89d2ac5 chore: Remove Azure SDK dependency (#198)
- e0dd658 chore: Commenting and Renaming (#194)
- a0f963e chore: Refactor to implement model plugin - part 5 (#191)
- c1151fd chore: Refactor to implement model plugin - part 4 (#190)
- e0f3a35 chore: Refactor to implement model plugin - part 3 (#189)
- 2d0c593 chore: put models under presets/models - Part 1 (#187)
- 4571883 chore: Refactor to implement model plugin - part 2 (#185)
- e5e6515 chore: Refactor to implement model plugin - part 1 (#183)
- e8615bd chore: bump actions/github-script from 6 to 7 (#176)
- cc8950e chore: bump step-security/harden-runner from 2.6.0 to 2.6.1 (#177)
v0.1.0
v0.1.0 - 2023-11-15
Changelog
Features 🌈
- c679ff9 feat: Add Preset Tests (#162)
- 97224eb feat: Add llama text completion (#149)
- b25218e feat: Improve README (#136)
- 2a01182 feat: adding workspace validation (#133)
Bug Fixes 🐞
- b97a288 fix: add custom template (#168)
- 5a20f66 fix: null ptr fix (#169)
- 4131348 fix: Add image path env vars (#167)
- 6c4fa45 fix: add tolerations for custom pod template (#161)
- 30f2442 fix: nil pointer in validatecreate (#160)
- b53e3ae fix: fix typo and add linguist override (#153)
- c997133 fix: add node affinity based on resource label (#152)
- 05f1a41 fix: use correct file header (#147)
- 50feff7 fix: use spaces (#145)
- b79e622 fix: change the annotation key for enabling LB type service (#142)
- c16cca4 fix: casing (#141)
- 1d854e3 fix: SocketError torch timeout (#135)
Maintenance 🔧
- 27fa335 chore: Revise main README.md and Preset README.md (#144)
- 7257b1f chore: bump google.golang.org/grpc from 1.55.0 to 1.56.3 (#139)
- 77a540a chore: bump golang.org/x/net from 0.15.0 to 0.17.0 (#140)
- 808db34 chore: remove unused files (#138)
- 478e45e chore: revise README.md to describe Kaito architecture (#132)
Security Fix 🛡️
- f3be1d5 security: Fix CVE-2023-44487 (#166)
- 18c6f34 security: Fix CVE-2023-3955 and CVE-2023-5528 (#165)
Testing 💚
- fa60f47 test: added ut for createTemplateInference (#164)
- e3cc935 test: added uts for createResource, getResource and generateMachineManifest (#163)
- 2c807b0 test: Add webhook e2e tests (#158)
- 1d11fa1 test: added ut for resources/nodes.go (#157)
- 098ecd1 test: add ut for resources/manifests.go (#154)
- e51a7e4 test: added ut for machine.go (#156)
- bde3886 test: added ut for applyWorkspaceResource (#155)
- 34c730c test: add ut for CreatePresetInference (#151)
- 87a4f10 test: added ut for deleteWorkspace (#150)
- fd5bd65 test: added ut for getAllQualifiedNodes (#148)
- a97b699 test: added ut for apply inference with template (#143)
- a83d820 test: Add e2e tests for kaito (#137)
- a4e3cc2 test: added ut for applyInference with preset (#134)
- 7bb88c6 test: added test for ensureService (#131)
v0.0.1
v0.0.1 - 2023-11-02
Changelog
Features 🌈
- 64c7a89 feat: Add accessmode, custom image, and image secret (#98)
- 874df03 feat: add PodTemplate support in workspace (#96)
- e912c57 feat: Added Falcon 40B Deployment (#90)
- 0223a50 feat: Include Falcon Deployment on KAITO (#80)
- 4df4266 feat: Auto image build (#67)
- ce26b36 feat: Make sure machine name is unique (#84)
- b067e76 feat: Add GPU plugins to chart (#83)
- d1780be feat: Remove DADI code (#77)
- 575c97c feat: Automated Preset Docker Image Building (#57)
- 4801723 feat: Add Deployments for E2E Tests (#60)
- a7eb78c feat: Benchmarking falcon inference (#52)
- 980c198 feat: Added Falcon Model Inference API (#51)
- e030ed1 feat: Merge kdm-preset-models into presets folder (#50)
- 94a7509 feat: Change CRD to support model access mode (#49)
- c84563d feat: Add statefulsets for distrbuted model inference (#48)
- a7ea650 feat: Add preset inference struct and support storage (#47)
- fd11734 feat: Add validation checks for immutable fields (#44)
- 87b69ec feat: Add validation webhook scaffolding code (#39)
- 2a5372d feat: generalize check resource status function (#8)
- cfe49a9 revert: "feat: Add skaffold code for webhook (#35)" (#36)
- 3490bcc feat: Add skaffold code for webhook (#35)
- b4aed40 feat: Add Inference conditions (#24)
- bab4ec1 feat: Merge machine conditions (#18)
- 98ef0e2 feat: Add default value if count not set (#16)
- dd21b97 feat: Add timeout for machine check status and return when SKU is not available (#13)
- 09ef00f feat: Implement inference deployments for llama2 (#12)
- 54b302e feat: Add more status for machine workflow (#8)
- 4c86b7a feat: Add workflow status (#7)
- d23a5d7 feat: Implement Machine creation (#6)
Bug Fixes 🐞
- 098f032 fix: conditional run matrix (#129)
- b556e6a fix: small nits (#128)
- 91a0de5 fix: GitHub runner name needs quotes (#127)
- d0ba4d5 fix: inference fault tolerance (#108)
- ab74683 fix: rename webhook secret name and remove leader election (#126)
- 0a5a9a2 fix: remove unused RBAC permission for workspace create/delete (#124)
- 9495158 fix: remove unused RBAC permissions and port (#122)
- 4e5c9b1 fix: change chart to use MCR and revise README (#121)
- 5356685 fix: revise README.md to add cluster name to chart (#120)
- ee48f1e fix: ensure check inference workload status (#112)
- 1815baa fix: aimodelsregistry ACR Push on Dispatch (#110)
- 1d8a582 fix: Use runner 0 for git checkout (#109)
- ffd96e6 fix: tag name (#105)
- e81db11 fix: filepath for inference file (#104)
- 7447413 fix: change key to kaito from gpu-provisioner (#101)
- 1844db2 fix: change Preset and Template to pointer fields (#99)
- 47305bc fix: fix a bug in counting existing nodes (#88)
- 14377f2 fix: make sure workspace gc cleans machines (#87)
- 2cc68b8 fix: Update nvidia-device-plugin-ds.yaml (#86)
- 8fdb4b2 fix: Update e2e-preset-test.yml (#85)
- b8ab612 fix: Update README.md (#82)
- 084d633 fix: Update README.md (#81)
- 530dfaf fix: Update the go package module (#58)
- 7a3c057 fix: rename the pipeline script (#43)
- ea150fc fix: optimize the use of crd status (#41)
- 7a8932e fix: Only update status when not matching (#40)
- a3e06b0 fix: include torch params in inference command (#38)
- 24f4433 fix: Update kdm.io_workspaces.yaml (#7)
- 3c69ae1 fix: add storage for llama 70b (#1)
Code Refactoring 💎
Continuous Integration 💜
- 3921d44 ci: Support multi-arch images (#115)
- bf57498 ci: Add gpu-provisioner to the e2e pipeline (#103)
- 4957501 ci: Add KAITO workspace e2e pipeline (#89)
- 64e43d2 ci: Add Publish to ACR GitHub workflow (#42)
- 0f14f67 ci: Fix checkout action
- 6f02a50 ci: Add unit tests Github workflow (#37)
- 004417a ci: Update create tag and upload artifacts (#32)
- e8df268 ci: Add goreleaser.yml
- 2a1b7fc ci: Add upload/download artifact to get pr context (#23)
- 469a5ad ci: Remove the workflow name from action-download-artifact action
- 2809d6b ci: Create helm chart pipeline and update release workflow (#5)
Maintenance 🔧
- 413de7a chore: revise README to include quick start (#123)
- d42fa0b chore: bump google.golang.org/grpc from 1.55.0 to 1.56.3 (#106)
- 7f6db90 chore: update the latest gpu-provisioner chart (#114)
- 6a37070 chore: revise README.md to refine installation steps (#111)
- 969d773 chore: update README for new installation guidence (#107)
- b1a1dc1 chore: follow Azure OSS codes (#102)
- 8da899b chore: fix update conflict and clean up logs (#100)
- bb3fa2d chore: put everything in examples folder (#94)
- 1be4d68 chore: rename accessmode type in CRD (#95)
- 5eb9a59 chore: cleanup presets folder (#93)
- 82b7398 chore: change k8sresources to resources (#92)
- 71b59bb chore: Refactor calculating number of new machines (#91)
- 421cc6d chore: bump thehanimo/pr-title-checker from 1.4.0 to 1.4.1 (#66)
- 64daef4 chore: bump golang.org/x/net from 0.10.0 to 0.17.0 (#61)
- 05df63b chore: bump step-security/harden-runner from 2.5.1 to 2.6.0 (#59)
- 735226f chore: Organize preset code (#53)
- b78ae6d chore: bump goreleaser/goreleaser-action from 4 to 5 (#45)
- 9e8082a chore: bump docker/login-action from 2.2.0 to 3.0.0 (#46)
- 5c17266 chore: bump actions/checkout from 3 to 4 (#3)
Revert Change ◀️
- b6aa55d revert: "release: update manifest and helm charts for v0.0.1 " (#33)
- a3e8ce2 revert: "release: update manifest and helm charts for v0.0.1 " (#29)
- a3b70a5 revert: "release: update manifest and helm charts for v0.0.1" (#27)
- a92ed79 revert: "release: update manifest and helm charts for v0.0.1" (#25)