quarkus-langchain4j
quarkus-langchain4j copied to clipboard
quarkiverse
Quarkus Langchain4j extension
Replaces #634. This PR aims to show how to use multiple model providers (Vertex AI Gemini, Azure OpenAI) and how they can access models using OIDC access tokens authorized by...
Currently the `OllamaStreamingChatLanguageModel` doesn't support tool integration. Let's add it. I'll take this and work on it.
https://docs.quarkiverse.io/quarkus-langchain4j/dev/ollama.html#quarkus-langchain4j-ollama_quarkus-langchain4j-ollama-chat-model-enabled What does "enabling a model" means? What happens when it's "off" https://docs.quarkiverse.io/quarkus-langchain4j/dev/ollama.html#quarkus-langchain4j-ollama_quarkus-langchain4j-ollama-enable-integration When set to false, what impact does that have? Disableing requests means nothing works? the LLM is...
We currently have a custom impl. I don't remember the exact details why we did that, but maybe the upstream impl would work for us now. Maintaining our own has...
The idea is to understand if there is a way to change the parameters passed to the LLM at runtime. Today the prompt can be changed using `@UserMessage` and `@V`:...
In this PR, the idea is to implement the semantic cache as discussed in #637. The idea is to allow the developer to use the _@CacheResult_ annotation to decide which...
Just a very rough draft for now, to continue the discussion. Added one sample where the LLM summarizes a git diff that I took from this repository. Maybe the samples...
At the Brno JUG meeting, there was a suggestion to allow easier experimenting with RAG in the sense of automatically ingesting files that get added to Easy RAG's document directory...
The current RAG model for pgvector is to store the documents in their own table. In my application my source documents already have a table: ```java @Entity public class Talk...
Perhaps this could be a dropdown menu that contains some useful system messages, and when you select one of them, it will get copied to the 'system message' field. Some...