Pricing

Extreme is the plan the product is built around.

Standard is the hosted desk. Expert adds audits, GitHub, and a team. Extreme adds the priority queue, daily watchdog, missions, pull requests, and the API. Limits are enforced on the server.

Standard

$30 / month

The hosted desk for a few active codebases.

3 active projects
40 engine runs per month
Uploads up to 5 MB and 200 files
Scan, diagnose, safe repair, code review, and code generation
Hosted Llama 3.2 3B on Workers AI (@cf/meta/llama-3.2-3b-instruct)
Bring your own OpenAI, Anthropic, OpenRouter, or Groq key
14-day run history
1 seat
Email support tickets

Choose Standard

Expert

$50 / month

Audits, GitHub connect, and a weekly watchdog.

10 active projects
200 engine runs per month
Uploads up to 25 MB and 2,000 files
Scan, diagnose, safe repair, code review, and code generation
Hosted Llama 3.1 8B on Workers AI (@cf/meta/llama-3.1-8b-instruct)
Bring your own OpenAI, Anthropic, OpenRouter, or Groq key
90-day run history
3 team seats
Security audit for secrets, wildcard dependencies, and a missing lockfile
Connect a GitHub repository with your own token
Scheduled watchdog scans, weekly
Email support tickets

Choose Expert

Best value

Extreme

$80 / month

The plan with the priority queue, missions, and pull requests.

20 active projects
1,000 engine runs per month
Uploads up to 100 MB and 8,000 files
Scan, diagnose, safe repair, code review, and code generation
Hosted Llama 3.3 70B on Workers AI (@cf/meta/llama-3.3-70b-instruct-fp8-fast)
Bring your own OpenAI, Anthropic, OpenRouter, or Groq key
365-day run history
10 team seats
Security audit for secrets, wildcard dependencies, and a missing lockfile
GitHub pull-request auto-fix from a safe repair
Scheduled watchdog scans, daily or weekly
Priority job queue
Mission runs: diagnose, safe repair, review, and generate
API keys for the project API
Priority support tickets

Choose Extreme

What each plan enforces

Capability	Standard	Expert	Extreme · recommended
Active projects	3	10	20
Runs / month	40	200	1,000
Upload	5 MB · 200 files	25 MB · 2,000 files	100 MB · 8,000 files
History	14 days	90 days	365 days
Seats	1	3	10
Hosted model	Llama 3.2 3B	Llama 3.1 8B	Llama 3.3 70B
Security audit	No	Yes	Yes
GitHub connect	No	Yes	Yes
GitHub pull request	No	No	Yes
Watchdog	No	every 168h+	every 24h+
Priority queue	No	No	Yes
Mission runs	No	No	Yes
API keys	No	No	Yes
Priority support	No	No	Yes

Local Ollama use stays free. The first paid invoice can be refunded within 14 days. See the refund policy. Cancel in the Stripe customer portal. A downgrade that leaves you over the project limit blocks new projects and new runs until you archive the extra ones. Past due and canceled subscriptions stay readable.