/ labAI · Tools
Multi-provider tokenizer + cost estimator
Counts tokens with the cl100k_base BPE tokenizer (used by recent OpenAI models) and applies the pricing tables from the cuesheet PyPI library to estimate per-call cost across families. Claude and Gemini counts use the same BPE as a close approximation, since neither provider ships an in-browser tokenizer; the relative order is reliable.
stackgpt-tokenizercuesheet pricing
178 chars · 0 input tokens (cl100k_base; Claude and Gemini counts land within ~5%)
Cost = (input × in-price + output × out-price). Both per million tokens.
| Model | In / 1M | Out / 1M | Cost per call | Per 1K calls |
|---|---|---|---|---|
Claude Opus 4.7 claude-opus-4-7 | $5.00 | $25.00 | $0.006400 | $6.40 |
Claude Sonnet 4.5 claude-sonnet-4-5 | $3.00 | $15.00 | $0.003840 | $3.84 |
Claude Haiku 4.5 claude-haiku-4-5 | $1.00 | $5.00 | $0.001280 | $1.28 |
GPT-5 gpt-5 | $1.25 | $10.00 | $0.002560 | $2.56 |
GPT-4o gpt-4o | $2.50 | $10.00 | $0.002560 | $2.56 |
GPT-4o mini gpt-4o-mini | $0.15 | $0.60 | $0.000154 | $0.15 |
Gemini 2.5 Pro gemini-2.5-pro | $1.25 | $10.00 | $0.002560 | $2.56 |
Gemini 2.5 Flash gemini-2.5-flash | $0.30 | $2.50 | $0.000640 | $0.64 |
pricing source · litellm / model_prices_and_context_window.jsonfetched 2026-05-21cached 24h · auto-refreshes