v0.4.1 - 2025-01-03
This release includes these major new features:
- Add support for Qwen2.5 Coder model.
- Add support for using LoRA adapter with vLLM runtime.
- Add support for config file for vLLM runtime.
- Add demoui for openai api service.
Changelog
Features 🌈
- 0c2f4a8 feat: qwen2.5 coder Proposal (#801)
- d9dc364 feat: add inference config api (#791)
- 84f58a3 feat: ensureServices for RAG engine (#776)
- a994f4b feat: update preset images (#785)
- be3620b feat: add qwen preset test (#788)
- d399599 feat: add qwen coder model (#783)
- 42f9ebc feat: support config file for vllm runtime (#780)
- 7da6586 feat: add demoui for openai api (#777)
- 83f25cd feat: support LoRA adapters for vllm runtime (#774)
- b099c66 feat: RAGEngine update and validation (#747)
- c3be988 feat: Add build pipeline for RAG Controller (#772)
- 9a2f8d6 feat: Add build pipeline for RAG Service (#770)
Bug Fixes 🐞
- 82451cb fix: unstable testing order causing flaky test (#799)
- 2c1d5bf fix: don't switch current working git branch when determining model changes (#789)
- 24eb89b fix: machine and nodeclaim can not supported at the same time (#769)
- e11c6d4 fix: Update Ragengine Service Dockerfile
- 857c373 fix: RAG service Dockerfile path patch (#767)
- d6f8602 fix: chart tpl when rendering feature gate flag (#760)
Documentation 📘
- 9dd17a3 docs: Add invite link to the Kaito community slack (#792)
- 5d5e342 docs: add LLMs chat template documentation for end-users (#782)
- d783693 docs: update README for new release (#762)
Maintenance 🔧
- 0d061fa chore: bump github.com/onsi/gomega from 1.34.2 to 1.36.2 (#794)
- d97b290 chore: bump goreleaser/goreleaser-action from 6.0.0 to 6.1.0 (#688)
- d96cdd5 chore: bump step-security/harden-runner from 2.10.1 to 2.10.2 (#729)
- 2b18896 chore: bump actions/setup-go from 5.1.0 to 5.2.0 (#781)
- ea9fed4 chore: bump codecov/codecov-action from 5.1.1 to 5.1.2 (#793)
- c96b10e chore: bump golang.org/x/net to 0.33.0 (#786)
- 53319e0 chore: bump github.com/onsi/ginkgo/v2 from 2.20.2 to 2.22.0 (#717)
- eb2bff2 chore: bump thehanimo/pr-title-checker from 1.4.2 to 1.4.3 (#728)
- 47bdfd8 chore: bump codecov/codecov-action from 5.0.7 to 5.1.1 (#766)
- 59e377f chore: update phi3.5 example resource (#763)
- cbd58d9 chore: switch buildkit image to mcr registry (#761)