Skip to main content
VePrompts

Token-to-Cost Converter

Enter input and output tokens, pick a model, and instantly see cost per 1K, cost per 1M, and total USD across the full VePrompts pricing database.

Last updated: 2026-06-13

Share
Quick presets:

Tokens sent to the model in your prompt.

Tokens the model generates in its response.

OpenAI · 128,000 context · up to 16,384 output tokens

Selected model pricing

Input $2.50/1M
Output $10.00/1M

Total estimated cost

$0.0075

for 1,500 tokens

$0.0050
per 1K tokens
$5.00
per 1M tokens
$0.0025
input cost
$0.0050
output cost

Cost comparison

Blended input + output price per 1M tokens. The selected model is highlighted.

Auto Router
$-2000000.00/1M
Body Builder (beta)
$-2000000.00/1M
OpenRouter: Fusion
$-2000000.00/1M
Pareto Code Router
$-2000000.00/1M
Venice: Uncensored (free)
$0.00/1M
Google: Gemma 4 26B A4B (free)
$0.00/1M
Google: Gemma 4 31B (free)
$0.00/1M
Google: Lyria 3 Clip Preview
$0.00/1M
Google: Lyria 3 Pro Preview
$0.00/1M
LiquidAI: LFM2.5-1.2B-Instruct (free)
$0.00/1M
OpenAI: GPT-4o selected
$12.50/1M

How LLM token pricing works

Input and output tokens are priced separately

Most providers charge a different rate for the tokens you send (input) and the tokens the model generates (output). Output tokens are usually more expensive because generation requires compute. This converter handles both sides automatically.

Prices are shown per 1M tokens

Provider rate cards list prices per 1 million tokens. A price of $2.50 means you pay $2.50 for every 1,000,000 tokens processed. The converter also normalizes this to a per-1K-token rate so you can reason about small prompts.

Why token count matters

API bills, context window limits, and rate limits are all measured in tokens. Knowing the cost of a single request lets you forecast monthly spend and compare models before you integrate them.

Shareable calculations

The URL updates as you type. Copy the address bar to share an exact cost estimate with your team, or bookmark a common workload. Your inputs never leave your browser.