Flagship Claude Sonnet 4.5
$0.0000 per request
- Input
- $0.0000
- Output
- $0.0000
$3/M in · $15/M out
VIBE CODING / AI-ASSISTED DEVELOPMENT
AI Prompt Cost Calculator
Paste your prompt, pick an expected response length, and see an estimated dollar cost per request across Claude, GPT, and Gemini flagship and smaller models.
- Updates live — no submit button
- Uses approximate published per-million-token pricing
- Compares a flagship and a cheaper model per family
Expected response length
Input: 0 tokens (varies per model family)
Smaller / cheaper Claude Haiku 4.5
$0.0000 per request
- Input
- $0.0000
- Output
- $0.0000
$1/M in · $5/M out
Flagship GPT-4.1
$0.0000 per request
- Input
- $0.0000
- Output
- $0.0000
$2/M in · $8/M out
Smaller / cheaper GPT-4.1 mini
$0.0000 per request
- Input
- $0.0000
- Output
- $0.0000
$0.4/M in · $1.6/M out
Flagship Gemini 2.5 Pro
$0.0000 per request
- Input
- $0.0000
- Output
- $0.0000
$1.25/M in · $10/M out
Smaller / cheaper Gemini 2.5 Flash
$0.0000 per request
- Input
- $0.0000
- Output
- $0.0000
$0.3/M in · $2.5/M out
Prices are approximate and based on each vendor's publicly listed per-million-token rates at the time this tool was last reviewed. Vendors change pricing without notice — verify current rates on the vendor's own pricing page before using this for budgeting or billing.
How it works
Paste your prompt and pick a response length
Paste the prompt text you plan to send, then choose an expected response length from short replies to long multi-file output.
Tokens are estimated per model family
Input tokens reuse the same family-specific approximation as the Token & Context Window Counter; output tokens use your selected preset directly.
Cost is calculated from published per-million-token rates
Each model card multiplies input and output tokens by that model's approximate published price per million tokens, shown separately, then summed into a total estimated cost per request.
Continue learning: Why AI coding credits run out · How to manage and version prompts in an AI product
Frequently asked questions
Are these prices current?
They are approximate, based on each vendor's publicly listed pricing at the time this tool was last reviewed. Vendors change prices without notice, so verify current rates on the vendor's own pricing page before using this for budgeting.
Why are there two models per family?
Most vendors offer a flagship model and a smaller, cheaper model with the same general capability tier. Comparing both shows the cost trade-off between quality and price for your expected usage.
How is the input token count calculated?
Using the same heuristic per-family approximation as the AI Token & Context Window Counter tool — not an exact vendor tokenizer. See that tool for the method.
Does this include the cost of a full conversation history?
No. It estimates the cost of one prompt plus one response. Multi-turn conversations re-send prior context on each turn, which this single-request estimate does not account for.
Can I use this for a monthly budget estimate?
Multiply the per-request total by your expected request volume for a rough monthly figure, but treat it as directional — actual usage, caching discounts, and pricing changes will shift the real number.
How We Compare
| Feature | MVPHub | OpenAI Pricing Calculator | LLM Price Check |
|---|---|---|---|
| Compares Claude, GPT, and Gemini in one view | Included | Not included | Limited |
| Estimates from your actual pasted prompt | Included | Not included | Not included |
| Shows flagship vs. smaller model trade-off | Included | Not included | Limited |
| No signup required | Included | Included | Included |
OpenAI's own pricing calculator covers GPT models only. Cross-vendor pricing trackers exist but rarely let you paste your actual prompt to estimate input tokens directly. MVPHub combines both in one live calculator.
Related tools
Embed this tool
Add this tool to your site with the canonical iframe below. It remains hosted and maintained by MVPHub.