Compare AI Model API Costs
Instantly calculate and compare pricing for 20+ LLMs. Input your token usage and see exact costs — per request, daily, monthly, and yearly.
Select Model
Input: —
Output: —
Context: —
Provider: —
Usage Inputs
Cost Breakdown
Per Request
—
1 API call
Daily
—
—
Monthly
—
30 days
Yearly
—
365 days
Input cost/day: —
Output cost/day: —
Quick Presets
Select Models to Compare (up to 3)
tokens
tokens
req/day
Comparison Results
| Model | Input $/1M | Output $/1M | Per Request | Monthly | Yearly | vs Cheapest |
|---|
| Model | Provider | Input ($/1M tokens) | Output ($/1M tokens) | Context | Input cost |
|---|
* Prices are per 1 million tokens. API prices as of April 2026. Prices may vary by tier and region. Free/open-weight models shown at $0.
FAQ — AI Model Pricing
How is AI model API pricing calculated?
+
AI model API pricing is typically charged per million tokens processed. Costs are split between input tokens (your prompt) and output tokens (the model's response). Multiply your token count by the per-million rate to get your cost. For example, GPT-4o charges $2.50 per million input tokens and $10 per million output tokens. A rough rule: 1 token ≈ 0.75 English words.
Which AI model is the cheapest in 2026?
+
In 2026, the cheapest commercial AI models include GPT-4.1-nano ($0.10/$0.40 per 1M tokens), Mistral Small ($0.10/$0.30), and Gemini 2.5 Flash ($0.15/$0.60). For open-weight models, Gemma 4 and Llama 4 Scout can be run for free or at very low cost via providers like Together AI at ~$0.18 per million tokens.
What is the difference between input and output token pricing?
+
Input tokens are the text you send to the model (your prompt, context, system instructions). Output tokens are the text the model generates back. Output tokens are usually 4–10x more expensive than input tokens because generating text requires significantly more computation than reading/processing it.
How much does it cost to run an AI agent for a month?
+
The cost depends heavily on the model and usage volume. A typical AI agent making 1,000 requests/day with 2,000 input tokens and 500 output tokens would cost about $6/month with GPT-4o-mini, $45/month with Claude Sonnet 4, or under $1/month with GPT-4.1-nano. Use the calculator above to get exact numbers for your use case.
Is Claude more expensive than GPT-4o?
+
It depends on the tier. Claude Haiku 3.5 ($0.80/$4) is cheaper than GPT-4o ($2.50/$10). Claude Sonnet 4 ($3/$15) is slightly more expensive than GPT-4o for output tokens. Claude Opus 4 ($15/$75) is significantly more expensive. For most production use cases, GPT-4o-mini and Claude Haiku offer similar cost-performance profiles at the budget end.