Quick answer
A Claude app subscription is not the same as the cost of a Claude bot. If you use Claude in its app, the listed U.S. prices are Free: $0; Pro: $20 per month, or $200 billed upfront for a year; Max: $100 per month; and Ultra: $200 per month. A bot built with the Anthropic API is billed separately according to its model and input and output token use. Neither the app subscription nor the API bill necessarily covers the full cost of building and operating a bot. The prices and rates below were checked November 12, 2026. (Claude pricing; Ultra plan details)
Claude app subscription prices
| Plan | Listed U.S. price | What to know |
|---|---|---|
| Free | $0 | Free app plan. |
| Pro | $20 monthly, or $200 upfront for a year (shown as $17/month) | Pro is for Claude’s chat experience; it does not include ordinary API usage through Claude Console. |
| Max | $100/month | Includes $100/month in Claude Platform API credits for eligible subscribers. |
| Ultra | $200/month | Includes $200/month in Claude Platform API credits for eligible subscribers. The no-five-hour-limit benefit applies to verified accounts. |
The Max and Ultra API credits are not an automatic discount in the estimate below: they require an eligible active plan and claiming the credits by linking a Claude Console organization. Check the credit terms before budgeting with them. (Claude Pro plan; API credit terms; Ultra plan details)
Claude API prices for a bot
Anthropic lists API prices in U.S. dollars per million tokens (MTok), with separate input and output rates. Input tokens are what the model reads; output tokens are what it generates. These standard rates provide a starting point for estimating a bot’s model usage:
| API model | Input / 1M tokens | Output / 1M tokens | Official positioning |
|---|---|---|---|
| Claude Haiku 5.5 | $0.10* | $0.50* | High-volume, latency-sensitive tasks such as classification, extraction, and routing. |
| Claude Sonnet 5.5 | $2 | $10 | Everyday coding, agent, and enterprise workloads; Anthropic describes it as a combination of speed and intelligence. |
| Claude Opus 5.5 | $4 | $20 | Long-running agentic coding and knowledge work. |
| Claude Fable 5.2 | $8 | $40 | Demanding reasoning and long-horizon agentic work. |
*Haiku’s listed $0.10 input and $0.50 output rates apply to prompts up to 100,000 tokens; higher rates apply above that threshold. (Anthropic API pricing; Haiku 5.5 overview; Sonnet 5.5 overview; Opus 5.5 overview; Fable 5.2 overview)
Example: estimate API usage for 10,000 monthly bot replies
Suppose a text bot handles 10,000 requests per month, averaging 1,000 input tokens and 200 output tokens per request. That is 10 million input tokens and 2 million output tokens per month. Assuming each prompt stays below Haiku’s 100,000-token threshold and using standard rates:
- Haiku 5.5: (10 × $0.10) + (2 × $0.50) = $2/month
- Sonnet 5.5: (10 × $2) + (2 × $10) = $40/month
- Opus 5.5: (10 × $4) + (2 × $20) = $80/month
- Fable 5.2: (10 × $8) + (2 × $40) = $160/month
These are illustrative API token-cost estimates, not measured bot bills or all-in operating prices. They assume the stated average tokens per request and do not include platform charges or other pricing features. Longer conversations can increase input usage if earlier context is sent again. To estimate your own workload, use: (monthly input tokens ÷ 1,000,000 × input rate) + (monthly output tokens ÷ 1,000,000 × output rate). (Anthropic API pricing)
Which model fits your bot?
- Consider Haiku for high-volume, latency-sensitive tasks such as classification, extraction, or routing. Treat it as a candidate and test output quality on representative requests before choosing it for production. (Haiku 5.5 overview)
- Consider Sonnet for everyday or well-scoped work, including general-purpose replies, coding, and data analysis. Anthropic describes it as built for everyday coding, agent, and enterprise workloads. (Sonnet 5.5 overview)
- Consider Opus or Fable for more demanding work when their documented focus on long-running agentic work or demanding reasoning is relevant to your bot. Their per-token rates are higher, so test them against your task and budget. (Opus 5.5 overview; Fable 5.2 overview)
A higher API rate alone does not establish that a model is better for your bot. Test your actual prompts and success criteria, then compare quality, latency, and cost on the same workload.
What the API estimate does—and doesn’t—cover
The example estimates Anthropic model-token usage only. It is not a quote for building or running a complete chatbot. Depending on your setup, you may also need to budget for hosting, bot or automation software, a messaging channel, storage, integrations, development, or maintenance. Which costs apply—and what they cost—depends on your implementation, so there is no supported single all-in price here.
Budget route: If you just want to chat with Claude yourself, compare the app’s Free and Pro options; Pro is a separate subscription and does not include ordinary Claude Console API usage. If you need a bot in your own product or workflow, estimate API token usage and add only the platform and operating costs your setup actually requires. If you have an eligible Max or Ultra plan, check the API-credit terms before counting those credits against a production estimate. (Claude Pro plan; API credit terms)
Sources
Plans & pricing | Claude by Anthropic
What is the Pro plan? | Claude Help Center
Monthly API credits for Max, Ultra, and Team plans | Claude Help Center
What is the Ultra plan? | Claude Help Center
Pricing | Claude Platform Docs
Claude Haiku 5.5 | Claude Platform Docs
Claude Sonnet 5.5 | Claude Platform Docs