Max Plans
The Max plans are the high-volume Command Code tiers for power users running agents all day. Both unlock every model - open-source and premium, including Claude Opus, Claude Fable, and Fugu Ultra - with no restrictions.
See every plan side by side on Pricing & Limits, or pick a plan from Studio > Billing.
| Plan | Price/mo | Credits/mo | Included LLM Usage | Models |
|---|---|---|---|---|
| Max 10× | $100 | $150 | ~230K requests | Every model, no restrictions |
| Max 20× | $200 | $300 | ~370K requests | Every model, no restrictions |
Both include more credits for the same cost as raw API pricing. Credits reset monthly, top up at model cost, roll over, and never expire.
Want premium models on a smaller budget? The Pro plan is $20/month for $80 - everything except Claude Opus, Claude Fable, and Fugu Ultra.
$100/month · $150 of credits · ~230K requests. Ten times the Pro budget for power users running Command Code throughout the workday.
- Every model, including Claude Opus, Fable, and Fugu Ultra. No model restrictions.
- ~230K requests/month on a balanced mix (~35% open-source / ~65% premium).
- Best for engineers who keep an agent running all day across multiple projects.
$200/month · $300 of credits · ~370K requests. The top tier - maxed-out daily usage at the highest request volume.
- Every model, no restrictions, including Claude Opus, Fable, and Fugu Ultra.
- ~370K requests/month on a balanced mix.
- Best for the heaviest users and small teams who want one big personal pool.
Max 10× and Max 20× include every model - the full open-source + premium catalog with no restrictions, including Claude Opus, Claude Fable, and Fugu Ultra.
The live table on this page lists every included model with its per-token rates. Switch it to All models for the full catalog including retired models.
Switch any time with /model. The full list also lives on Available Models.
The request numbers above are estimates, not ceilings. They assume a balanced mix - roughly ~35% open-source / ~65% premium.
- A typical request. ~700–1K input tokens, ~125–200 output tokens, plus ~42K–56K cache reads on average.
- Long conversations cost more. Each turn re-reads the full context, so use
/clearor a new session for unrelated tasks. - Track the real number. The Usage page in Studio always reflects the actual price charged per request.
Model the math yourself with the calculator on Pricing & Limits.
- Resets monthly. Your request budget refreshes at the start of every billing cycle.
- Top up automatically. Buy extra credits at model cost. They roll over. They never expire.
- Deals apply automatically. Discounts and usage multipliers apply to every request, including extra credits.
- Deals stack on top of your credits. Your credit bonus lands first, then any active model deal applies to those credits - a model at 50% off makes them go twice as far. See Deals for what's running right now.
- Max 10× - for power users running Command Code all day.
- Max 20× - for maxed-out daily usage at the highest request volume.
Want premium models on a modest budget? The Pro plan is $20/month. On a budget, or only need open-source models? Start with the Go plan at $1/month. Want raw pay-as-you-go API access to every model with no markup? See the Provider API. Teams and organizations: contact support@commandcode.ai about Team & Enterprise plans.
- Install:
npm i -g command-code - Sign in and start coding with
cmd - Pick Max 10× or Max 20× from Studio > Billing or the pricing page
Full comparison and FAQs live on Pricing & Limits.
The Max plan has API access. You call the Provider API endpoints with your Command Code API key, and usage is metered against your Max credits and the models included above.
Subscribe
Pick the Max plan from Studio > Billing or the pricing page. Every plan except the Go plan has API access.
Create an API key
The same key authenticates the CLI and the API. Create one from your API keys page in Studio.
Call the API
Point any OpenAI or Anthropic compatible client at https://api.commandcode.ai/provider/v1 and send your first request. Premium models take the Anthropic Messages shape on /v1/messages, open models take the OpenAI shape on /v1/chat/completions.
Provider API Example Request
Read the Provider API quickstart for more details.
We still recommend the Command Code CLI: harness engineering makes the same credits go further - the Read tool keeps ~25 billion junk tokens a month out of your context, tool call repairs are free on every plan, and ~98% cache hit rates mean most of every request bills as cheap cache reads.
What models are available on the Max plans?
What models are available on the Max plans?
Every model, with no restrictions - the full open and premium catalog, including Claude Opus, Claude Fable, and Fugu Ultra. Switch any time with /model in interactive mode. Browse them all on Available Models, with per-token rates on Pricing & Limits.
Should I pick Max 10× or Max 20×?
Should I pick Max 10× or Max 20×?
Max 10× is $100/month for $150 of credits (~230K requests) - built for engineers who keep an agent running all day. Max 20× is $200/month for $300 of credits (~370K requests) - the top tier for the heaviest users. Both unlock every model; the only difference is how much you get.
What are the usage limits on the Max plans?
What are the usage limits on the Max plans?
On Max 10×, $45 of usage in any 5 hours and $90 in any 7 days; on Max 20×, $90 and $180. Windows roll from your first request and reset on their own - see Usage Limits. Extra pay-as-you-go credits have no usage limits.
Can I buy extra credits?
Can I buy extra credits?
Yes, any time. Buy pay-as-you-go credits at model cost - they unlock every model, roll over, never expire, and skip the usage limits entirely.
Can I use the Max plans via API?
Can I use the Max plans via API?
Yes. Every plan except the Go plan has API access - you use the same Provider API the Provider plan uses (OpenAI Chat Completions and Anthropic Messages endpoints, one key for the CLI and the API), so you can integrate Max with any other agent. We still recommend the Command Code CLI, because phenomenal harness engineering makes the same credits go further: our Read tool keeps ~25 billion junk tokens a month out of your context, tool call repairs are free on every plan, and we run ~98% cache hit rates so most of every request bills as cheap cache reads.
Are the Max plans available worldwide?
Are the Max plans available worldwide?
Yes. Open-source models run on infrastructure in the US, EU, and Singapore for reliable access worldwide.
Can I enforce zero data retention?
Can I enforce zero data retention?
Yes. Most models are ZDR by default - agreements are renewed monthly and can take time for brand-new models. You can also enforce ZDR on every request (which can change model prices based on the provider, as explained in Pricing & Limits). Run the CLI with CMD_ZDR=1 (for example, CMD_ZDR=1 cmd) to enforce zero data retention and no prompt training on every request.
Can I upgrade, downgrade, or cancel any time?
Can I upgrade, downgrade, or cancel any time?
Yes. Change or cancel your plan from Studio > Billing at any time. Upgrades take effect immediately and start both usage windows fresh; downgrades take effect at your next billing cycle. Extra credits you have bought stay with your account - they roll over and never expire.