Developers waste hours comparing AI model pricing across scattered provider docs. One API. Every model. Always current.
AI model pricing is fragmented across dozens of provider docs, constantly changing, and impossible to compare at a glance. Building cost-aware LLM routing means maintaining your own pricing spreadsheet. We solved that.
Everything you need to make cost-informed model decisions
Pass token counts, get exact cost breakdowns including cache pricing. Perfect for budgeting and billing.
Compare any models head-to-head: pricing, context windows, capabilities. Make data-driven routing decisions.
Consistent response format across all endpoints. Filter by provider, capability, or category. Pagination built in.
Native Model Context Protocol support. Let your AI assistant query pricing directly via tool calls.
Runs on Cloudflare Workers at 300+ edge locations. Sub-50ms responses globally. Zero cold starts.
Pricing data updated with every deploy. Tracks all major providers including cache and batch pricing tiers.
Copy, paste, see results in seconds
curl "https://api.lazy-mac.com/llm-pricing/api/v1/compare?models=gpt-4o,claude-sonnet-4-20250514,gemini-2.5-pro&tokens=1000"curl "https://api.lazy-mac.com/llm-pricing/api/v1/calculate?model=claude-sonnet-4-20250514&input_tokens=100000&output_tokens=5000"
// Response:
{
"ok": true,
"data": {
"model": "claude-sonnet-4-20250514",
"input_tokens": 100000,
"output_tokens": 5000,
"input_cost": 0.30,
"output_cost": 0.075,
"total_cost": 0.375,
"currency": "USD"
}
}curl "https://api.lazy-mac.com/llm-pricing/api/v1/models?provider=anthropic&capability=vision"24 providers and growing
Start free. Scale when you need to.
Let your AI query pricing directly. Our MCP endpoint exposes list_models, compare_models, and calculate_cost as native tools. Drop the URL into Claude Desktop, Cursor, or any MCP-compatible client.
POST https://api.lazy-mac.com/llm-pricing/mcp