Agenta
LLMOps platform for prompt management, LLM evaluation, and observability. Build, evaluate, and monitor production-grade LLM applications with collaborative prompt engineering.
High commit volume, frequent releases and growing activity.
Commit activity in the top 1% of tracked projects; popularity more than 66%. How this is calculated.
Commits, last 11 months
| Stars | 4.8k |
|---|---|
| Latest release | v0.121.4 · 27 Sept 2026 |
| Repo updated | 27 Sept 2026 |
| Licence | MIT |
| Built with | Docker |
Is Agenta actively maintained?
Agenta is under active development: 15000 commits landed over the last 11 months, averaging roughly 2223 commits a month in the most recent quarter.
Activity is accelerating: the last quarter carried 88% more commits than the one before it.
The most recent tagged release, v0.121.4, shipped within the last month — a current, installable version exists today.
Agenta ranks #6 of 14 in Generative Artificial Intelligence (GenAI), placing it mid-table for maintenance activity.
What Agenta actually does
Agenta is a self-hosted LLMOps platform designed to help teams manage prompts, evaluate language model outputs, and observe application behaviour in production. It provides a centralised workspace for collaborative prompt engineering and testing. The software is released under the MIT licence.
Best fit: Engineering teams building production applications on top of large language models who require collaborative prompt management and systematic evaluation.
Worth knowing: Evaluating and observing LLM workflows can generate significant data volumes, meaning you need to monitor storage and performance scaling carefully as traffic grows.
Deployment notes
Because the application runs via Docker, you will typically need to configure persistent volumes for data storage and place the container behind a reverse proxy to handle TLS termination.
Links
Agenta as a replacement for
Alternatives to Agenta
Projects in the same categories, ordered by health score.
Modern design AI chat framework supporting multiple AI providers, one click install MCP Marketplace and Artifacts / Thinking.
User-friendly AI Interface, supports Ollama, OpenAI API.
Run your AI models locally and generate images and audio (alternative to OpenAI and Claude).
Enhanced ChatGPT-compatible AI chat interface supporting multiple AI providers, with multi-user auth, message search, and plugin support.
Chat UI that works with any LLM. It comes loaded with advanced features like agents, web search, RAG, MCP, deep research, Connectors to 40+ knowledge sources, and more.
All-in-one desktop & Docker AI application with built-in RAG, AI agents, No-code agent builder, MCP compatibility, and more.