Launched November 11, 2025
Agenta

Agenta

Open-source prompt management & evals for AI teams

Subscription2,343 impressions#4 of its week

Agenta is an open-source LLMOps platform for building reliable AI apps. Manage prompts, run evaluations, and debug traces. We help developers and domain experts collaborate to ship LLM applications faster and with confidence.

Agenta is an open-source LLMOps platform for building, managing, and automating AI agents with prompt management, evaluations, and trace debugging.

for
AI developers, product teams, and domain experts building LLM applications
pricing
freemium
license
open source
Agenta-AI/agenta 4.8k 682TypeScriptupdated 18 days ago
works withSlackTelegramGitHubNotionHubSpotLinearGoogle DriveDiscord

Key features

6 features of Agenta
  • Unlimited agents & workflows — Create unlimited agents, projects, schedules, and event triggers on any plan.
  • Agent runs & tracing — Each interaction creates a trace; runs are billed per 10,000 beyond plan limits.
  • Multi-model support — Connect OpenAI, Anthropic, Gemini, Mistral, Groq, MiniMax, Together AI, Azure, AWS Bedrock, Google Vertex, or custom endpoints.
  • 1000+ integrations via MCP — Integrate with Slack, Telegram, GitHub, Notion, HubSpot, Linear, Google Drive, and many more.
  • Role-based access & SSO — Control permissions, roles, and single sign-on for enterprise governance.
  • Open-source self-hosted option — MIT-licensed Docker/Helm deployment with full control over data and infrastructure.

Use cases

  • Automated code review comments on every pull request
  • Research and draft SEO articles directly from Slack or Telegram
  • Scheduled KPI monitoring and automated reporting
  • Customer-support ticket triage via chat agents
  • Cross-tool data aggregation and report generation

Agenta pricing

  • Hobby$02 team members · 5,000 agent runs/month · 1-week trace retention · Community support
  • Pro$29/monthUnlimited users · 10,000 agent runs/month · 1-month trace retention · Unlimited evaluations
  • Business$299/monthTeam roles & RBAC · SSO & SOC 2 Type II · 3-month trace retention · Priority support
  • Open source$0Unlimited users & projects · Self-hosted deployment · Bring your own models · Role-based access

Agenta FAQ

What counts as an agent run?+

Any tracked execution—sending a message, creating an annotation, or triggering a schedule—creates one agent run and a corresponding trace.

Are AI model costs included in the plan price?+

No. You connect your own model provider credentials and pay the provider directly for usage.

Can I self-host Agenta?+

Yes. The open-source edition is MIT-licensed and can be deployed via Docker Compose or Helm charts on your infrastructure.

What trace data retention does each Cloud plan offer?+

Hobby retains traces for 1 week, Pro for 1 month, Business for 3 months, and Enterprise is configurable.

Is there a free tier for teams?+

The Hobby plan is free forever for up to two team members and includes 5,000 agent runs per month.

Do unused agent runs roll over to the next month?+

No. Agent run allowances reset at the start of each billing period.

Summarized by DevHunt from agenta.ai · Sep 28, 2026. Details may change; check the official site.