Choosing the right AI model is a cost/quality tradeoff. This table compares every major model's pricing so you can make informed decisions. Prices are per million tokens.
| Model | Provider | Input/1M | Output/1M | Context | Speed (tok/s) |
|---|---|---|---|---|---|
| GPT-4o | OpenAI | $5.00 | $15.00 | 128K | ~95 |
| GPT-4o mini | OpenAI | $0.15 | $0.60 | 128K | ~130 |
| GPT-4 Turbo | OpenAI | $10.00 | $30.00 | 128K | ~40 |
| GPT-3.5 Turbo | OpenAI | $0.50 | $1.50 | 16K | ~150 |
| o1 | OpenAI | $15.00 | $60.00 | 200K | ~30 |
| o1-mini | OpenAI | $3.00 | $12.00 | 128K | ~65 |
| Claude 3.5 Sonnet | Anthropic | $3.00 | $15.00 | 200K | ~80 |
| Claude 3.5 Haiku | Anthropic | $0.80 | $4.00 | 200K | ~120 |
| Claude 3 Opus | Anthropic | $15.00 | $75.00 | 200K | ~25 |
| Gemini 1.5 Pro | $3.50 | $10.50 | 2M | ~70 | |
| Gemini 1.5 Flash | $0.075 | $0.30 | 1M | ~180 | |
| Gemini 2.0 Flash | $0.10 | $0.40 | 1M | ~200 | |
| Mistral Large | Mistral | $4.00 | $12.00 | 128K | ~55 |
| Mistral Small | Mistral | $1.00 | $3.00 | 128K | ~110 |
| Llama 3.1 70B | Meta (Together) | $0.88 | $0.88 | 128K | ~75 |
Assuming average 500 input tokens and 1,500 output tokens per prompt:
| Model | Per prompt | Per 1K prompts | Per 100K prompts |
|---|---|---|---|
| GPT-4o | $0.025 | $25 | $2,500 |
| GPT-4o mini | $0.001 | $1 | $100 |
| Claude 3.5 Sonnet | $0.024 | $24 | $2,400 |
| Gemini 1.5 Flash | $0.0005 | $0.50 | $50 |
Paste your real prompt into WeighMyPrompt to see exact token counts and cost for every model side by side. Adjust the output ratio for your use case. Free, private, no sign-up.