avante.nvim icon indicating copy to clipboard operation
avante.nvim copied to clipboard
verified publisher icon yetone

feat: Enhanced Model Provider Support and Configuration Flexibility For Rag Service

Open doodleEsc opened this issue 7 months ago • 2 comments

rag_service was limited to only Ollama and OpenAI as model providers, and it didn't allow separate configuration of language models and embedding models. LlamaIndex offers integration with multiple providers, which has enabled me to add support for different vendors. Additionally, I've modified the configuration file format to accommodate the need to separately configure language models and embedding models.

doodleEsc avatar May 20 '25 09:05 doodleEsc

‌‌‌Thank you for your PR. Could you help resolve the pre-commit issue?

yetone avatar May 22 '25 02:05 yetone

all fixed, please take a look.

doodleEsc avatar May 22 '25 08:05 doodleEsc

This PR is stale because it has been open 14 days with no activity. Remove stale label or comment or this will be closed in 10 days.

github-actions[bot] avatar Jun 06 '25 02:06 github-actions[bot]

LGTM!

yetone avatar Jun 06 '25 15:06 yetone

@yetone @doodleEsc @cinuor1 Thank you for the enhancement. Could you please push the required image 0.11 Currently we have only the previous version and the rag_service is not working anymore https://quay.io/repository/yetoneful/avante-rag-service?tab=history

bmihovski avatar Jun 06 '25 17:06 bmihovski