Media
1 item
Description
See what your AI chats actually cost. Live token and cost estimates on ChatGPT, Claude, and Gemini, across 9 models. The same prompt can cost 25× more on one model than another — but chat UIs never show you a price. CostPerPrompt adds a live cost badge to ChatGPT, Claude, and Gemini: • This prompt: estimated tokens in, and estimated total cost including a typical reply. • This conversation: total tokens of the messages visible on the page. • Cheapest alternative: what the same prompt would cost on the cheapest model, and the spread vs the model you're on. • Popup comparator: paste any prompt and see its estimated cost on nine current models at once, cheapest first, from budget mini models to frontier flagships. • Export: save a session as Cost Per Correct JSON, compatible with the tokenecon cost-per-correct workflow. Prices are public list prices (USD per 1M tokens, checked Oct 2026). Token counts use a transparent on-device heuristic (±10–15% vs a provider tokenizer) — built for model comparison, not billing. Powered by tokenecon, the open-source token-economics toolkit: https://pypi.org/project/tokenecon/ 100% local. No server, no analytics, no account. Session snapshots stay in your browser and can be cleared any time.