Cost Predictor
What does your prompt cost?
Paste any prompt below — instantly see the price across all 12 providers in our gateway. The cheapest option is often 10-100× less than the default everyone uses.
💰 Switch from Claude Opus 4.7 to GPT-5 mini → save $0.009089 per call (99.2% cheaper)
At 10,000 calls/month: $90.89 saved. At 1M calls/month: $9089.00 saved.
Cost across 14 models
Sorted cheapest first. All prices include OneAPIKey's transparent 20% markup. Direct provider prices are slightly lower; what you save is unified billing, smart routing, and 12-provider failover.
| Model | Provider | Input cost | Output cost | Total cost | Context | Privacy |
|---|---|---|---|---|---|---|
👑 GPT-5 mini gpt-5-mini | OpenAI | $0.000001 | $0.000072 | $0.000073 | 128K | ✓ |
Codestral codestral | Mistral | $0.000002 | $0.000072 | $0.000074 | 32K | 🛡️ |
Llama 3.3 70B (Groq) llama-3-3-70b | Groq | $0.000006 | $0.000094 | $0.000100 | 128K | ✓ |
DeepSeek V3 deepseek-v3 | DeepSeek | $0.000002 | $0.000132 | $0.000134 | 128K | ⚠️ |
DeepSeek R1 deepseek-r1 | DeepSeek | $0.000005 | $0.000262 | $0.000267 | 66K | ⚠️ |
Claude Haiku 4.5 claude-haiku-4-5 | Anthropic | $0.000008 | $0.000480 | $0.000488 | 200K | 🛡️ |
Gemini 2.5 Pro gemini-2-5-pro | Google Gemini | $0.000013 | $0.000600 | $0.000613 | 2000K | ✓ |
Mistral Large mistral-large | Mistral | $0.000021 | $0.000720 | $0.000741 | 128K | 🛡️ |
GPT-5 gpt-5 | OpenAI | $0.000027 | $0.001200 | $0.001227 | 256K | ✓ |
Command R+ command-r-plus | Cohere | $0.000027 | $0.001200 | $0.001227 | 128K | ✓ |
Claude Sonnet 4.6 claude-sonnet-4-6 | Anthropic | $0.000032 | $0.001800 | $0.001832 | 200K | 🛡️ |
Sonar Pro sonar-pro | Perplexity | $0.000032 | $0.001800 | $0.001832 | 200K | ✓ |
Grok 4 grok-4 | xAI | $0.000032 | $0.001800 | $0.001832 | 256K | ✓ |
Claude Opus 4.7 claude-opus-4-7 | Anthropic | $0.000162 | $0.009000 | $0.009162 | 1000K | 🛡️ |
Use this prediction in your app
POST to our public API. No auth required.
curl -X POST https://oneapikey.app/v1/predict-cost \
-H "Content-Type: application/json" \
-d '{
"prompt": "Your prompt text here",
"output_tokens": 100
}'Stop overpaying for AI.
OneAPIKey routes your traffic to the right model automatically — based on rules you define.
Start free