SourceAnt

Simple pricing, honest usage

Self-host the open-source core for free. When you want the managed cloud, it is one flat team price.

Self-host

$0

open source, forever

Run the open-source reviewer on your own infrastructure. Full control over your data.

  • AI code review on your GitHub
  • Self-hosted, your keys, your servers
  • Community support
  • MIT-licensed core
View on GitHub
Most popular

Managed

$25

per team / month + AI usage

The full cloud app, all four pillars. We run the models, and you pay for AI usage as you go.

  • Everything in Self-host, hosted for you
  • Context and memory, knowledge graph
  • Context-aware review quality
  • AI usage billed as you go
  • Unlimited seats per team
  • Email support
Start free

Bring your own keys

$50

per team / month, flat

A predictable flat price with no usage bills from us. Connect your own model keys and keep AI spend and data on your side.

  • Everything in Managed
  • Connect your own OpenAI, Anthropic, or other keys
  • Flat monthly price, no usage metering from us
  • Your AI spend and data stay with your provider
  • Unlimited seats per team
  • Email support
Start free

Start on SourceAnt Cloud

We are onboarding teams now. Leave your work email and we will get you set up.

Questions

What does "per team" mean?

One flat price covers your whole team. We do not charge per seat.

Managed vs bring your own keys?

Managed is $25 per team plus AI usage as you go, and we run the models for you. Bring your own keys is a flat $50 per team with no usage bills from us: you connect your own provider keys and pay them directly. Managed is cheaper to start; bring your own keys is predictable and keeps your AI spend and data on your side, which tends to win at heavier usage.

Is there really a free option?

Yes. The core reviewer is open source, so self-host it for free forever. The paid plans add the managed cloud, the knowledge graph, and context memory.

Which AI tools does it work with?

Any tool that speaks MCP. Context and graph queries are served over a standard interface, so you are not locked to one assistant.