The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Claude Code’s /usage command separates session usage into input, output, cache-read, and cache-write tokens, broken down by model. Those counts describe different parts of requests; the cost shown in Claude Code is an estimate, not an authoritative API bill. For API billing, check the Claude Console Usage page.
What the four token counts mean
In an agentic coding session, a request can contain more than your latest message. Claude Code sends model instructions, conversation context, and—when tools are involved—tool definitions, tool calls, and tool results. Anthropic notes that tool requests are priced on the total input sent, including the tools parameter and tool_use and tool_result blocks. (Anthropic API pricing)
![Claude AI for Beginners Bible: [5 in 1] The Ultimate Guide to Automate Your Work, Save Hours Every Week, and Use AI for Real-World Results](https://m.media-amazon.com/images/I/415+fSJacsL._SL160_.jpg)
- Input tokens: Material sent to the model, including relevant conversation content and tool-related payloads.
- Output tokens: Text and other content generated by the model. API pricing treats output separately from input.
- Cache-write tokens: Prompt content stored in the prompt cache. A write is charged when content is first stored.
- Cache-read tokens: Cached prompt content retrieved by a later request. A read is charged separately from ordinary input.
Cache reads and writes are input-side usage, not output tokens, and neither category should be assumed to be free. Anthropic’s current general API pricing rules list cache writes at 1.25× base input for a five-minute cache or 2× for a one-hour cache, and cache reads at 0.1× base input for most listed models. Model-specific exceptions and other pricing modifiers can apply, so consult the live pricing page for current rates.
How to check tokens in Claude Code
- In a Claude Code session, run
/usage. The/costcommand is an alias. - Read the Session block for token totals by model, including input, output, cache reads, and cache writes. The displayed cache counts come from cache-token fields in the API response. Claude Code’s guide says its cache statistics cover the main conversation, not subagents; availability and display details can change, so check the current command documentation.
- To see how much of the active context window is being used, run
/context. This view can show context-heavy tools and capacity warnings, but it is not a billing statement.
Claude Code’s cost guide documents the session usage and cost display.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallWhy the displayed cost can differ from your bill
Claude Code calculates its local API session cost from token counts at list prices, unless an organization-managed modelPricing table applies. The guide labels this figure an estimate and directs API users to the Claude Console Usage page for authoritative billing. The CLI’s --max-budget-usd limit also uses a client-side estimate, which can differ from the bill. (cost guide; CLI usage documentation)
How to interpret that figure depends on how you authenticate:
- API users: Treat the local session cost as an estimate; use the Claude Console Usage page for the billing record.
- Pro and Max subscribers: Usage is included in the subscription, so the session cost figure is not a measure of a separate API bill.
- Gateway-routed sessions: The gateway credential and upstream provider determine who is billed. Anthropic says an active gateway credential replaces the subscription login for those requests, which are billed per token to the owner of the forwarded credential. See the LLM gateway documentation.
Token totals, context usage, and comparisons
/usage answers how many tokens the session used and shows its estimated cost. /context answers how much of the active context window is in use. These measures are related, but they are not interchangeable: context consumption is not a billing statement, and a session usage total is not a context-capacity gauge. See the Claude Code command reference.
Character count or word count cannot reliably reproduce the tokenization of a complete Claude Code request. For actual usage, rely on the session or API usage fields rather than a general character-to-token conversion.
When comparing sessions or providers, keep the model, input and output totals, cache reads and writes, authentication route, and source of the cost figure consistent. For API price comparisons, also account for the current model rate, cache duration, provider, and applicable pricing modifiers. A subscription usage bar is not directly comparable to a per-token API invoice.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
