LLM price calculator

List prices for 4,600+ models from the LiteLLM catalog — the snapshot promptry bills against.

# estimate a workload

Model

Input tokens / call

Output tokens / call

Calls / month

Input from cache (%)

# cost

Per call—

Per 1,000 calls—

Per month—

input — output — rate —

These are provider list prices. Real bills swing with prompt-cache hits, retries, and tool loops — promptry records the number you're actually charged →

Model	Input $/1M	Cached $/1M	Output $/1M

Prices in USD per 1,000,000 tokens. Cached = input served from the provider's prompt cache. A blank cell means the provider doesn't publish that rate.

Stop estimating. Start measuring.

A calculator guesses. promptry logs every real call — cost, tokens, the cached-token split, and which prompt spent it — from a one-line import. One SQLite file, no account.

The 30-second setup What is promptry?

# every call priced automatically from promptry.openai import OpenAI client = OpenAI() client.chat.completions.create(…)