OverpayingForAIPricing desk

Lesson 2 of 3 · 7 min read · Beginner

The real cost: whichever model you plug in

OpenClaw's own price is $0. Its running cost is entirely the token bill of the model behind it — and that bill can vary by more than 10x depending on which model you choose.

In this lesson you will

  • Understand why OpenClaw's cost tracks the model you connect, not the gateway
  • Compare a rough monthly cost across several real model backends
  • Use the fill-in exercise to estimate cost for your own message volume

Every message your OpenClaw bot reads and every reply it writes passes through an LLM API call, billed per token like any other API use. The gateway adds no markup — it's a routing layer, not a paid product — so your bill is exactly the model provider's own per-token rate times how much the bot actually talks.

Worked example: ~500 input + ~150 output tokens per reply, 1,000 replies$ per 1,000 repliesLlama 3.3 70B0.1 $0.10/$0.32 per 1M tokensDeepSeek V30.28 $0.26/$1.03 per 1M tokensGPT-5 Mini0.43 $0.25/$2.00 per 1M tokensHermes 4 405B0.95 $1.00/$3.00 per 1M tokensClaude Haiku 4.51.25 $1.00/$5.00 per 1M tokens
Figure 1.A worked example using this site's own catalogue rates, assuming ~500 input and ~150 output tokens per reply — a busy server bot can easily run 1,000+ replies a month.

Your turn

Estimate your own OpenClaw monthly bill

At roughly [MESSAGES_PER_DAY] replies/day using [MODEL] (about [INPUT_RATE] in / [OUTPUT_RATE] out per 1M tokens), expect around [ROUGH_MONTHLY] a month.

Hint: 50 replies/day × 30 days = 1,500 replies/month — scale the $/1,000-replies figures above by 1.5x for that volume.

Knowledge check

Your OpenClaw bot is quiet most of the month but suddenly does 20,000 replies during one busy week. What happens to your bill on a per-token API backend?

Lesson FAQ

How much does OpenClaw cost to run?

OpenClaw itself is free and open-source. The ongoing cost is entirely the token bill of whichever LLM you connect it to — a worked example on this site puts a busy bot's cost anywhere from about $0.10 to $1.25 per 1,000 replies depending on the model chosen.

Is there a way to run OpenClaw for free?

Self-hosting an open-weight model (Llama, Hermes) removes the per-token API bill, trading it for your own compute cost instead — genuinely free if you already have idle hardware, but not literally $0 if you're renting a GPU to do it.

Finished reading?

Mark it done to track your progress through the course.

Save progress across devices

Get a private link that restores your lessons on any device. Email is optional and only used to send you the link.

Compare, calculate, decide

If our calculators helped you cut down on hidden AI wallet leaks, thanks for using them. A tiny fraction of your savings is what keeps our pricing indexes updated daily.

Not sure which AI is cheapest for your use case? Find out in 30 seconds — no signup required.

AI cost intelligence

Stop overpaying for AI tools

Join the OverpayingForAI list for pricing updates, cheaper alternatives, and practical buying guidance.

Now tracking 50+ AI tools, models, platforms, subscriptions, coding tools, and automation products.

We use your email only for OverpayingForAI updates. Unsubscribe anytime.