LLM Cost Calculator
Estimate and compare API costs across OpenAI, Anthropic, Google, Meta, xAI, DeepSeek and more. Enter your usage below or pick a quick preset.
Quick compare for:
Top 12 Cheapest Models
Monthly total (USD) — sorted by cost
Average Cost by Provider
Avg across all models (USD/month)
Input vs Output Cost Split
Top 10 cheapest — stacked (USD/month)
- Input
- Output
All Models(46 results)
500 input tokens · 1,000 output tokens · 100 calls/month
| Provider | Model | Context | Input / 1M tokens | Output / 1M tokens | Per Call | Monthly Total |
|---|---|---|---|---|---|---|
| OpenAI | BestGPT-5 nano | 400k | $0.0500 | $0.4000 | $0.00043 | $0.0425 |
| Gemini 2.0 Flash | 1M | $0.1000 | $0.4000 | $0.00045 | $0.0450 | |
| Gemini 2.5 Flash Lite | 1M | $0.1000 | $0.4000 | $0.00045 | $0.0450 | |
| OpenAI | GPT-4.1 nano | 1M | $0.1000 | $0.4000 | $0.00045 | $0.0450 |
| DeepSeek | DeepSeek V3.2 | 128k | $0.2800 | $0.4200 | $0.00056 | $0.0560 |
| OpenAI | GPT-4o mini | 128k | $0.1500 | $0.6000 | $0.00068 | $0.0675 |
| Cohere | Command-R7B | 128k | $0.2500 | $0.7500 | $0.00087 | $0.0875 |
| xAI | Grok-3 Mini | 131k | $0.2500 | $0.7500 | $0.00087 | $0.0875 |
| Meta | Llama 4 Scout | 10M | $0.3000 | $1.0000 | $0.00115 | $0.1150 |
| Anthropic | Claude 3 Haiku | 200k | $0.2500 | $1.2500 | $0.00138 | $0.1375 |
| Cohere | Command-R | 128k | $0.5000 | $1.5000 | $0.00175 | $0.1750 |
| OpenAI | GPT-3.5 Turbo | 16k | $0.5000 | $1.5000 | $0.00175 | $0.1750 |
| OpenAI | GPT-4.1 mini | 1M | $0.4000 | $1.6000 | $0.00180 | $0.1800 |
| OpenAI | GPT-5 mini | 400k | $0.2500 | $2.0000 | $0.00213 | $0.2125 |
| DeepSeek | DeepSeek R1 | 64k | $0.5500 | $2.1900 | $0.00247 | $0.2465 |
| Kimi | Kimi K2 | 128k | $1.6700 | $1.6700 | $0.00250 | $0.2505 |
| Gemini 2.5 Flash | 1M | $0.3000 | $2.5000 | $0.00265 | $0.2650 | |
| Meta | Llama 3.3 70B | 128k | $0.7500 | $2.5000 | $0.00287 | $0.2875 |
| Anthropic | Claude 3.5 Haiku | 200k | $0.8000 | $4.0000 | $0.00440 | $0.4400 |
| OpenAI | o3-mini | 128k | $1.1000 | $4.4000 | $0.00495 | $0.4950 |
| OpenAI | o4-mini | 200k | $1.1000 | $4.4000 | $0.00495 | $0.4950 |
| xAI | Grok-3 | 1M | $1.5000 | $4.5000 | $0.00525 | $0.5250 |
| Anthropic | Claude Haiku 4.5 | 200k | $1.0000 | $5.0000 | $0.00550 | $0.5500 |
| Gemini 1.5 Pro | 2M | $1.2500 | $5.0000 | $0.00562 | $0.5625 | |
| Meta | Llama 4 Maverick | 1M | $1.5000 | $5.0000 | $0.00575 | $0.5750 |
| OpenAI | GPT-4.1 | 1M | $2.0000 | $8.0000 | $0.00900 | $0.9000 |
| OpenAI | o3 | 200k | $2.0000 | $8.0000 | $0.00900 | $0.9000 |
| Cohere | Command-A | 256k | $3.0000 | $9.0000 | $0.01050 | $1.0500 |
| Gemini 2.5 Pro | 2M | $1.2500 | $10.0000 | $0.01063 | $1.0625 | |
| OpenAI | GPT-5.1 | 1M | $1.2500 | $10.0000 | $0.01063 | $1.0625 |
| OpenAI | GPT-4o | 128k | $2.5000 | $10.0000 | $0.01125 | $1.1250 |
| OpenAI | GPT-5.2 | 400k | $1.7500 | $14.0000 | $0.01488 | $1.4875 |
| Anthropic | Claude 3.7 Sonnet | 200k | $3.0000 | $15.0000 | $0.01650 | $1.6500 |
| Anthropic | Claude Sonnet 4 | 200k | $3.0000 | $15.0000 | $0.01650 | $1.6500 |
| Anthropic | Claude Sonnet 4.5 | 200k | $3.0000 | $15.0000 | $0.01650 | $1.6500 |
| Anthropic | Claude Sonnet 4.6 | 200k | $3.0000 | $15.0000 | $0.01650 | $1.6500 |
| Anthropic | Claude Opus 4.6 | 200k | $5.0000 | $25.0000 | $0.02750 | $2.7500 |
| OpenAI | GPT-5 | 1M | $5.0000 | $30.0000 | $0.03250 | $3.2500 |
| OpenAI | GPT-4 Turbo | 128k | $10.0000 | $30.0000 | $0.03500 | $3.5000 |
| xAI | Grok-4 | 256k | $10.0000 | $40.0000 | $0.04500 | $4.5000 |
| Cohere | Command-R+ | 128k | $15.0000 | $45.0000 | $0.05250 | $5.2500 |
| OpenAI | o1 | 200k | $15.0000 | $60.0000 | $0.06750 | $6.7500 |
| Anthropic | Claude Opus 4 | 200k | $15.0000 | $75.0000 | $0.08250 | $8.2500 |
| OpenAI | o3-pro | 200k | $20.0000 | $80.0000 | $0.09000 | $9.0000 |
| OpenAI | GPT-5 pro | 400k | $15.0000 | $120.0000 | $0.12750 | $12.7500 |
| OpenAI | GPT-5.2 pro | 400k | $21.0000 | $168.0000 | $0.17850 | $17.8500 |
What is a token?
A token is roughly 4 characters or ¾ of a word in English. "Hello world" is about 3 tokens. Most LLMs price per 1,000 or 1,000,000 tokens processed.
What’s the difference between input and output tokens?
Input tokens are your prompt (instructions + context). Output tokens are the model’s reply. Output pricing is usually 3–5× higher than input.
Why does Image Creation show text-model costs?
Image generation APIs (DALL-E, Midjourney) are priced per image, not per token. This preset estimates the cost of using an LLM to write/refine image prompts before sending to an image API.
How accurate are these estimates?
Prices reflect publicly available API pricing as of early 2026. Actual costs may vary with volume discounts, caching, fine-tuning tiers, or regional pricing.