LLM API Cost Calculator
Paste a prompt and an expected output length to estimate the API cost across Claude, GPT, and Gemini models side by side. Prices are refreshed from a public model-pricing catalog every time the site is rebuilt.
Everything runs locally in your browser — your text is never sent to any server, only the price list is fetched ahead of time at build.
Loading tokenizer dictionary…
Applied to every model equally — there is no generated text to tokenize per model, so this is a single estimate you control.
Cost by model
| Model | Input tokens | Output tokens | $ / 1M in · out | Est. cost |
|---|---|---|---|---|
| Claude Opus 4.8 | 0 | 0 | $5 · $25 | $0.00 |
| Claude Sonnet 4.5 | 0 | 0 | $3 · $15 | $0.00 |
| Claude Haiku 4.5 | 0 | 0 | $1 · $5 | $0.00 |
| GPT-5.5 | 0 | 0 | $5 · $30 | $0.00 |
| GPT-5 | 0 | 0 | $1.25 · $10 | $0.00 |
| GPT-4.1 | 0 | 0 | $2 · $8 | $0.00 |
| GPT-4o | 0 | 0 | $2.5 · $10 | $0.00 |
| GPT-4 Turbo | 0 | 0 | $10 · $30 | $0.00 |
| GPT-3.5 Turbo | 0 | 0 | $0.5 · $1.5 | $0.00 |
| Gemini 3 Pro | 0 | 0 | $2 · $12 | $0.00 |
| Gemini 2.5 Flash | 0 | 0 | $0.3 · $2.5 | $0.00 |
Pricing data updated: 2026-08-19 · Source: OpenRouter public model catalog (reference pricing, not official vendor pricing).
Costs are estimates for rough comparison only — not for billing. Claude and Gemini token counts are heuristic estimates (±10–20%), and per-model chat-API overhead (message structure, system prompt, tool definitions) is not included. Always check the vendor's own pricing page before making a purchasing decision.
The tokenizer dictionary could not be loaded, so the GPT-based input token count is also an approximation.
Articles for this tool
Recent Articles
curl Options Cheat Sheet: What -X, -H and -d Actually Do
A reference for the curl options you meet in real API work: why -d already implies POST, how -d differs from --data-raw, the @ prefix that silently reads a file, single vs double quotes, and why -k and -L deserve more caution than they usually get.
Converting Between JSON Schema and Zod: How required Maps to .optional()
Inside a two-way converter that turns JSON Schema into a Zod schema and Zod code back into JSON Schema. Covers the inverted defaults between required and .optional(), the constraint mapping table, and how the Zod side is parsed without executing any code.
SQL Clause Order Reference: Why WHERE Can't See Your SELECT Alias
The order you write SQL clauses is not the order the database runs them. A reference for the logical execution order (FROM → WHERE → GROUP BY → HAVING → SELECT → ORDER BY → LIMIT), why aliases fail in WHERE, when to use WHERE vs HAVING, and the MySQL/PostgreSQL differences that bite.
Unified Diff Format Reference: Reading @@ Hunks in git diff Output
How to read the unified diff format git produces: what the four numbers in @@ -12,7 +12,9 @@ mean, why a one-character edit shows as a whole-line replacement, the whitespace and line-ending traps, \ No newline at end of file, combined @@@ diffs on merges, and rename detection via similarity index.
UTC to JST Reference: The 9-Hour Offset, Cheat Sheet & Timezone Pitfalls
Convert between UTC and JST (Japan Standard Time) with a cheat sheet. Covers what Z and +09:00 mean in ISO 8601, when JavaScript date parsing silently shifts by 9 hours, MySQL/PostgreSQL timezone behavior, and why GitHub Actions cron always runs in UTC.
CREATE TABLE Reference: MySQL vs PostgreSQL vs SQLite Types & Constraints
A cross-database CREATE TABLE (DDL) reference with cheat sheets for data types, auto-increment keys (AUTO_INCREMENT / IDENTITY / rowid), foreign key ON DELETE behavior, and the CHECK constraint that MySQL silently ignores.