agenta
agenta copied to clipboard
Agenta-AI
The all-in-one LLM developer platform: prompt management, evaluation, human feedback, and deployment all in one place.
Right now the add to test set button is in the input card in the playground: [https://share.cleanshot.com/4bDnn8pD](https://share.cleanshot.com/4bDnn8pD)  We would like to move it to the text...
Part of https://github.com/Agenta-AI/agenta-commons/pull/54
Add a button in the playground left to Add to test set show logs which shows the trace information for the call as a drawer in the same style done...
The goal of this issue is to optimize the identified queries (see the project description) in Atlas. The optimization can happen by identifying the beanie query, removing fetch_links, and instead:...
## Description This PR improves the get_config endpoint latency by 86.7% ### Related Issue Closes #1599 ### Dependency [commons_#53](https://github.com/Agenta-AI/agenta-commons/pull/53) ### Preview [config-latency-improved.webm](https://github.com/Agenta-AI/agenta/assets/55067204/098dfbf8-585c-4cce-be7d-6d5d6c166d53)
We had multiple bugs in llm_apps_service in the past. The logic for invoking the applications is quite complex. We need to have tests for the different scenarios and errors. We...
Creates multiple evaluations, sometimes after they start, the table get stuck in loading stage while the too calls take an eternity to finish: GET api/evaluations/6637ce6bf8f86b074d948eb1/status/ GET api/evaluations/app_id **GOALS:** * The...
From [SyncLinear.com](https://synclinear.com) | [AGE-176](https://linear.app/agenta/issue/AGE-176/write-documentation-for-get-prompt-in-the-sdk)
## Description When using Agenta in development mode, you are expected to install packages that are not directly related to the app you want to serve.  ### Expected Behaviour...
Currently, all API endpoints are stored in a single file. We need to refactor the endpoints into the `services` folder and group them based on functionality.
