Billing
Subscription Tiers
| Feature | Free | Cloud |
|---|---|---|
| Price | $0/month | $10/month |
| Repositories | 3 (public only) | Unlimited (public + private) |
| Runs per day | 10 | Unlimited |
| History retention | 30 days | 1 year |
| Shared agent memory | Yes | Yes |
| PR comments (via Action) | Yes | Yes |
| Included LLM credits | $5/month | $10/month |
| Credit top-ups | Yes | Yes |
| BYO API key | Yes | Yes |
Plans are set per organization. Each organization (personal or team) has its own plan, credit balance, and billing. Only organization admins can manage billing.
The Free tier is designed for open-source projects and individual developers getting started with the CLI (which works fully standalone). The Cloud tier unlocks unlimited usage and private repo support. Shared agent memory is available on both tiers — when authenticated, the agent's navigation plans (which files matter and how to verify each condition) are synced to the cloud and shared across your team, reducing costs and improving accuracy on subsequent runs.
PR comments are free on all tiers. The GitHub Action posts PR comments using the built-in GITHUB_TOKEN, so no paid plan is required for PR feedback.
LLM Credits
Every Axiomatic plan includes monthly LLM credits that provide access to the Claude API without needing your own API key. Credits are managed entirely server-side per organization - cost is calculated from the Anthropic API response, so usage tracking is accurate and tamper-proof.
| Plan | Included Credits | Top-ups |
|---|---|---|
| Free | $5/month | $10, $25, $100 packs |
| Cloud | $10/month | $10, $25, $100 packs |
Included credits auto-reset every 30 days from the start of your billing period. Both plans support purchasing additional credit packs, which expire 1 year from purchase.
Using LLM Credits
- Authenticate with
axm login - Run tests as usual with
axm run- noANTHROPIC_API_KEYneeded - The CLI automatically routes requests through the Axiomatic LLM proxy
- Check your balance anytime with
axm credits
Cost Estimation
Typical costs per test depend on the model:
| Model | Cost per test | Best for |
|---|---|---|
| Claude Haiku | $0.01-0.05 | Most tests, fast iteration |
| Claude Sonnet | $0.05-0.20 | Standard accuracy tests |
| Claude Opus | $0.20+ | Critical security and architecture audits |
Caching means you only pay when relevant code changes. A typical project with 10 tests costs a few cents per CI run after the initial analysis.
BYO API Key
You can always bypass platform credits by setting your own API key:
export ANTHROPIC_API_KEY=sk-ant-...
axm runWhen a local API key is set, requests go directly to Anthropic and no platform credits are consumed. This is useful for teams with existing Anthropic contracts or for usage beyond the included credit limits.