LLM API cost calculator
Enter your workload and compare monthly cost across models.
| Model | Per request | Per month | Remove |
|---|---|---|---|
| Claude Haiku 5.5 | $0.00035 | $105.00 | |
| GPT-6 Luna | $0.00035 | $105.00 | |
| Qwen3.8 Flash | $0.000385 | $115.50 | |
| DeepSeek V4.1 Flash | $0.0009 | $270.00 | |
| MiniMax-M3 | $0.0009 | $270.00 | |
| GLM-5.3-FlashX | $0.000995 | $298.50 | |
| Gemini 3.5 Flash Lite | $0.00155 | $465.00 | |
| Grok Build 0.1 | $0.002 | $600.00 | |
| Seed 2.0 Pro | $0.002 | $600.00 | |
| Gemini 3.8 Flash | $0.00263 | $787.50 | |
| DeepSeek V4 Pro | $0.0033 | $990.00 | |
| Muse Spark 1.3 | $0.00337 | $1,013 | |
| Grok 4.7 | $0.005 | $1,500 | |
| Qwen3.8 Max | $0.005 | $1,500 | |
| Mistral Medium 3.5 | $0.00525 | $1,575 | |
| Claude Sonnet 5.5 | $0.007 | $2,100 | |
| GPT-6.1 Sol | $0.007 | $2,100 | |
| Gemini 3.1 Pro Preview | $0.008 | $2,400 | |
| Kimi K3 | $0.0105 | $3,150 | |
| Claude Opus 5.5 | $0.014 | $4,200 | |
| GPT-6 Astra | $0.035 | $10,500 |
Cost = input tokens × input price + output tokens × output price, 30 days a month. Cached input uses the cache read price. Cache write fees are not included.
Runs in your browser. Your files never leave your device.
How the estimate works
For every model the calculator takes your tokens per request and requests per day, applies the model's input, output and cache read prices, and multiplies by 30 days.
If your prompts are longer than a model's long-context threshold, the higher long-context price is used. Models without a published input or output price are left out.