Is Caveman really free?

The agent and SDK are free and open source with no Caveman account. Caveman Cloud Free is $0 for one person and needs no card. You still pay your model provider for inference on your own keys.

Do I need a card for Free?

No. Without a card, the allowances are hard limits until next month. Add a card to continue past them at standard rates.

$300 a month, including $300 of usage every month at the same standard rates used for events, agent minutes, and routing. Included usage does not roll over, and there are no seats.

How does the Team trial work?

Team starts with a 7-day free trial. A card is added at checkout and nothing is charged during the trial, including usage. When it ends, the $300 fee and its $300 of included usage start, prorated for the rest of that month.

How is Enterprise priced?

Enterprise is an annual contract, priced on the AI spend Caveman manages and never per seat. It runs hosted or in your own cloud.

What happens at the billing limit?

Caveman stops recording that product until the month resets and never charges past the limit. Gateway calls keep going to your provider unchanged. For agent runs, no new run starts; for routing, no decision is made.

What counts as an event?

One recorded LLM call, span, or eval score is one event. A gateway call counts one event per started 64 KB it sends.

Do you mark up model tokens?

No. Token markup is 0%. Your provider bills you directly for model calls on your own keys. Caveman bills for events, agent minutes, routing decisions, the Team fee, and any package.

Does my code leave my machine?

The agent and SDK compress locally and keep the original bytes on your machine. Cloud sees traffic routed through it; a customer-owned install keeps that traffic in your own cloud account.

Caveman Pricing in 2026: Plans, Limits and Real Costs

Free is the best starting point for a solo founder: it costs $0 for one person and includes the Cloud features up to stated allowances. Choose Team at $300 / month only when you need a whole team, $300 of included usage, and controls unavailable on Free.

Verified 2026-11-02 · official pricing page · Caveman in the directory

Quick pricing verdict

Starting price: Free is $0 / month for one person.
Free plan: Yes. Free includes 100,000 events a month, 600 agent minutes a month, 25,000 routing decisions a month, and $10 of Caveman AI a month.
Best plan for most people: Free is best for a solo founder evaluating Caveman. No card is needed.
Plan to avoid first: Enterprise is priced on the AI spend Caveman manages under an annual contract, so it is not a practical starting point for an individual.
Biggest hidden cost: Your model provider still bills you directly for inference on your own keys, even though token markup is 0%. Usage past Free allowances is charged at standard rates.
Pricing checked: 2026-11-02, against the official page

Start with Caveman Cloud Free at $0 / month. It supports one person, needs no card to begin, and provides 100,000 events a month, 600 agent minutes a month, 25,000 routing decisions a month, and $10 of Caveman AI a month. Without a card, those allowances are hard limits.

Team costs $300 / month and includes $300 of usage every month. It adds semantic cache, project access control, audit log export, and content-sharing controls; SSO is available with the Scale package. Enterprise uses a custom annual contract.

Plans compared

Plan	Price	Best for	Key limits
Free	$0 / month	One developer trying everything Caveman does, or a solo founder evaluating the service.	100,000 events a month; 600 agent minutes a month; 25,000 routing decisions a month; $10 of Caveman AI a month; $50 on Free with a card as the default billing limit per product.
Team	$300 / month$300 of usage every month	Teams that want agent costs cut, with proof, plus project and content-sharing controls.	Includes $300 of usage every month, then standard rates; $150 on agent runs past the included usage; no seats; prepaid usage and discounts are set up by hand.
Enterprise	Custom / year	Organizations that need an annual contract and can deploy Cloud in their own cloud.	Annual contract; hosted or in your VPC; AWS or GCP with Kubernetes; your KMS and storage; usage and billing by contract.

Source: Caveman pricing page, checked 2026-11-02. See the archived page we read.

What each plan includes

Free

Includes: Every Cloud feature for one person, 100,000 events a month, 600 agent minutes a month, 25,000 routing decisions a month, and $10 of Caveman AI a month.
Missing: No included usage credit; no semantic cache, project access control, audit log export, content-sharing controls, SSO, SCIM, or deployment in your VPC. Prompt text and recorded requests may be used to improve Caveman's models.
Avoid if: You need a whole team, included usage, project access control, audit log export, or data-sharing controls turned off.
Verdict: Best starting point for a solo founder; add a card only when you need to continue past the allowances.

Team

Includes: $300 of usage every month, semantic cache, project access control, audit log export, content sharing off, and SSO with the Scale package. There are no seats.
Missing: SSO requires Scale · $750/mo. Included usage does not roll over. Usage, model calls, and other standard rates can still create additional charges.
Avoid if: You are evaluating Caveman alone and do not yet need team controls or included usage.
Verdict: Choose when the team and control requirements justify $300 / month.

Enterprise

Includes: An annual contract, deployment in your VPC with one Helm chart on AWS or GCP, your KMS and storage, signed releases, and Enterprise usage terms.
Missing: The page lists Enterprise usage and billing as By contract rather than publishing a price.
Avoid if: You need a self-serve monthly plan or a known fixed price.
Verdict: For organizations requiring an annual contract or in-VPC deployment.

Add-ons, overages, and surprise costs

Cost item	Trigger	Published amount	Billing behaviour
Model-provider charges	Model calls run on your own provider keys.	Not published	Your provider bills you directly for inference. Caveman's token markup is 0%, but provider charges still apply.
Additional LLM observability events	Monthly events exceed 100,000 on Free or Team.	From $0.00035	Free and Team are charged per event at the standard rate for the applicable monthly volume.
Additional agent-run minutes	Monthly minutes exceed 600 on Free or Team.	From $0.02	Free and Team are charged per minute at the standard rate for the applicable monthly volume.
Additional routing decisions	Monthly decisions exceed 25,000 on Free or Team.	From $0.001	Free and Team are charged per decision at the standard rate for the applicable monthly volume.
SSO package	SSO is required.	$750/mo	SSO is not included on Free or Team; the comparison lists Scale · $750/mo for Enterprise.

Source: Platform pricing docs, checked 2026-11-02. See the archived page we read.

Which plan should you choose?

Solo founder trying Caveman for the first time: Free — It is $0 / month for one person and needs no card to begin.
Developer who expects to exceed the Free allowances: Free with a card — You can continue at standard rates, with a default $50 billing limit per product unless you set your own.
Team wanting included usage and access controls: Team — It costs $300 / month, includes $300 of usage every month, and adds semantic cache, project access control, audit log export, and content sharing off.
Organization requiring in-VPC deployment or an annual contract: Enterprise — Enterprise supports deployment in your VPC with one Helm chart on AWS or GCP, your KMS and storage, and an annual contract.

Frequently asked questions

Is Caveman really free?: The agent and SDK are free and open source with no Caveman account. Caveman Cloud Free is $0 for one person and needs no card. You still pay your model provider for inference on your own keys.
Do I need a card for Free?: No. Without a card, the allowances are hard limits until next month. Add a card to continue past them at standard rates.
How is Team priced?: $300 a month, including $300 of usage every month at the same standard rates used for events, agent minutes, and routing. Included usage does not roll over, and there are no seats.
How does the Team trial work?: Team starts with a 7-day free trial. A card is added at checkout and nothing is charged during the trial, including usage. When it ends, the $300 fee and its $300 of included usage start, prorated for the rest of that month.
How is Enterprise priced?: Enterprise is an annual contract, priced on the AI spend Caveman manages and never per seat. It runs hosted or in your own cloud.
What happens at the billing limit?: Caveman stops recording that product until the month resets and never charges past the limit. Gateway calls keep going to your provider unchanged. For agent runs, no new run starts; for routing, no decision is made.
What counts as an event?: One recorded LLM call, span, or eval score is one event. A gateway call counts one event per started 64 KB it sends.
Do you mark up model tokens?: No. Token markup is 0%. Your provider bills you directly for model calls on your own keys. Caveman bills for events, agent minutes, routing decisions, the Team fee, and any package.
Does my code leave my machine?: The agent and SDK compress locally and keep the original bytes on your machine. Cloud sees traffic routed through it; a customer-owned install keeps that traffic in your own cloud account.

What we read

Every figure on this page was extracted from the vendor's own pricing page at the moment below. The captured text is one click away, so you can check our reading against what they publish today.

Source captured 2026-11-02 08:22 UTC · read the captured text

Pricing change log

What moved since we started tracking this page, newest first. A price that changes silently is the failure mode of every other pricing guide.

2026-11-02First published, extracted from https://caveman.so/pricing; archive snapshot 2026-10-18.