langchainrb

langchainrb copied to clipboard

feat(hugging_face): Add support for multiple models and dimensions

3

Hello! Thank you for this awesome gem! I'm currently playing with it and i've seen Hugging face ruby gem can now support custom embedding models: see: https://github.com/alchaplinsky/hugging-face/commit/bee85a443bd085acbc61ad8926afa574aadfc7a4 I've added the...

bricesanchez

Vertex AI authentication issues

2

I'm having issues authenticating with Vertex with APPLICATION_DEFAULT_CREDENTIALS. I found a simple fix though. FROM @authorizer = ::Google::Auth.get_application_default TO @authorizer = ::Google::Auth.get_application_default(scope: 'https://www.googleapis.com/auth/cloud-platform')

palladius

### Description Currently OpenAI is the only supported LLM by `Langchain::Assistant`. We'd like to expand the list of support LLMs. ### Tasks - [x] Add support for Google Gemini (https://github.com/andreibondarev/langchainrb/pull/513)...

andreibondarev

Bedrock support is missing new models. Is there a reason the Bedrock client needs to white label models as opposed to just exposing the client since the api is unified?

1

weilandia

Chat Thread API [New Feature]

6

At the moment I am using the chat API with Ollama which is requiring me to maintain an array of inputs and responses in my application code and pass the...

scarrick68

Stream chat response in Assistants

4

This PR is an attempt to make streaming working in Assistants. Feel free to create a new PR or push to the same branch if you want to try different...

onerinas

Create `Langchain::Tool::Tavily` that uses Tavily service to do search.

2

### Description The current search Tool utilizes the `google_search_results` gem which is a [very wonky experience](https://github.com/andreibondarev/langchainrb/blob/main/lib/langchain/tool/google_search/google_search.rb#L52-L127). We'd like to use something cleaner. What's been proposed was using Tavily AI. ###...

andreibondarev

Cannot specify role: system in LLM::Anthropic

5

# Description In OpenAI and Ollama, it is possible to specify `system`, `user`, and `assistant` as roles. But in Anthropic, only `user` and `assistant` can be specified, and an error...

kokuyouwind

fix #603 As I wrote in the Issue proposal, I took the role: system message and merged it with the top-level system parameter.

kokuyouwind

`Langchain::LLM::Base#chat()` accepts a new `response_schema:` parameter to force the response to adhere to JSON schema

4

This PR introduces a new parameter `response_schema:` to the LLM `chat()` methods that, using Function Calling, forces the response to adhere to a specific schema. Example usage: ```ruby json_schema =...

andreibondarev