Current Claude API prices and cost estimates with prompt caching and Batch API pricing.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Compare what the same workload costs on every Claude model, cheapest first. Give tokens per request and the number of requests (e.g. per user per month).
Input tokens per request
Output tokens per request
Use Batch API pricing (default false)
Number of requests (default 1)
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http claude-pricing https://agentwright.app/mcp
Compare what the same workload costs on every Claude model, cheapest first. Give tokens per request and the number of requests (e.g. per user per month).
| Argument | Type | Description |
|---|---|---|
| batch | boolean | Use Batch API pricing (default false) |
| requests | number | Number of requests (default 1) |
| input_tokens* | number | Input tokens per request |
| output_tokens* | number | Output tokens per request |
{
"type": "object",
"required": [
"input_tokens",
"output_tokens"
],
"properties": {
"batch": {
"type": "boolean",
"description": "Use Batch API pricing (default false)"
},
"requests": {
"type": "number",
"minimum": 0,
"description": "Number of requests (default 1)"
},
"input_tokens": {
"type": "number",
"minimum": 0,
"description": "Input tokens per request"
},
"output_tokens": {
"type": "number",
"minimum": 0,
"description": "Output tokens per request"
}
},
"additionalProperties": false
}Estimate the USD cost of Claude API calls from token counts. Handles prompt caching (cache writes and reads), the Batch API 50% discount and long-prompt pricing. Returns cost per request and in total.
| Argument | Type | Description |
|---|---|---|
| batch | boolean | Use Batch API pricing (default false) |
| model* | string | Model id, e.g. claude-fable-5-2, claude-opus-5-5, claude-sonnet-5-5, claude-haiku-5-5 |
| requests | number | Number of requests (default 1) |
| input_tokens* | number | Uncached input tokens per request |
| output_tokens* | number | Output tokens per request (includes thinking tokens) |
| cache_read_tokens | number | Tokens read from the prompt cache per request |
| cache_write_1h_tokens | number | Tokens written to the 1-hour prompt cache per request |
| cache_write_5m_tokens | number | Tokens written to the 5-minute prompt cache per request |
{
"type": "object",
"required": [
"model",
"input_tokens",
"output_tokens"
],
"properties": {
"batch": {
"type": "boolean",
"description": "Use Batch API pricing (default false)"
},
"model": {
"type": "string",
"description": "Model id, e.g. claude-fable-5-2, claude-opus-5-5, claude-sonnet-5-5, claude-haiku-5-5"
},
"requests": {
"type": "number",
"minimum": 0,
"description": "Number of requests (default 1)"
},
"input_tokens": {
"type": "number",
"minimum": 0,
"description": "Uncached input tokens per request"
},
"output_tokens": {
"type": "number",
"minimum": 0,
"description": "Output tokens per request (includes thinking tokens)"
},
"cache_read_tokens": {
"type": "number",
"minimum": 0,
"description": "Tokens read from the prompt cache per request"
},
"cache_write_1h_tokens": {
"type": "number",
"minimum": 0,
"description": "Tokens written to the 1-hour prompt cache per request"
},
"cache_write_5m_tokens": {
"type": "number",
"minimum": 0,
"description": "Tokens written to the 5-minute prompt cache per request"
}
},
"additionalProperties": false
}List current Claude API models with their prices in USD per million tokens: input, output, prompt-cache writes (5 min / 1 h), cache reads, and the Batch API discount.
No arguments.
{
"type": "object",
"properties": {},
"additionalProperties": false
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 10, 2026.