AI API Pricing Dataset: LLM prices as JSON, CSV and a free API

Current input, output, cached-input and batch prices for OpenAI, Anthropic Claude, Google Gemini, xAI Grok, DeepSeek, Mistral AI, Cohere, Moonshot AI (Kimi), Alibaba Qwen and Meta. Every record links to the provider's own pricing page and carries the date it was last checked. Unknown prices stay null, never 0.

curl -L https://aicostbudget.github.io/ai-api-pricing-data/api/v1/prices.json

Compare every model side by side → · Estimate your monthly bill · See recent price changes

Prices at a glance

<!-- price-table:start (generated by scripts/update_readme_prices.py; do not edit) -->

USD per 1M tokens, standard list price. 130 models tracked, last verified 2026-10-31.

| Provider | Active token-priced models | Lowest output price (input / output) | Highest output price (input / output) | | --- | ---: | --- | --- | | OpenAI | 14 | GPT-6 Luna $0.10 / $0.50 | GPT-5.5 Pro $30.00 / $180.00 | | Anthropic | 11 | Claude Haiku 5.5 $0.10 / $0.50 | Claude Mythos 5.1 $10.00 / $50.00 | | Google Gemini | 12 | Gemini 3.1 Flash-Lite $0.25 / $1.50 | Gemini 3.5 Transcribe Live $3.50 / $21.00 | | xAI | 6 | Grok 4.20 (Non-Reasoning) $1.25 / $2.50 | Grok 4.7 $2.00 / $6.00 | | DeepSeek | 2 | DeepSeek V4.1 Flash $0.30 / $1.20 | DeepSeek V4 Pro $1.32 / $3.96 | | Mistral AI | 7 | Ministral 3 3B $0.10 / $0.10 | Mistral Medium 3.5 $1.50 / $7.50 | | Cohere | 2 | Aya Expanse 32B $0.50 / $1.50 | Command R+ 08-2024 $2.50 / $10.00 | | Moonshot AI (Kimi) | 3 | Kimi K3.1 Pro $2.00 / $10.00 | Kimi K3.1 Pro Highspeed $4.00 / $20.00 | | Alibaba Qwen | 4 | Qwen Flash $0.05 / $0.40 | Qwen3.8 Max $2.00 / $6.00 | | Meta | 1 | Muse Spark 1.3 $1.25 / $4.25 | same model |

Batch, cached-input, cache-write, long-context and non-token prices are in the full data.

<!-- price-table:end -->

Click a provider for its full price list, or a model for its JSON record. If this saves you a scrape, a ⭐ helps other developers find it. To hear about price changes, subscribe to the price-change Atom feed.

Get the data

| Format | Where | | --- | --- | | JSON | api/v1/prices.json | | CSV | api/v1/prices.csv | | One model | api/v1/models/<provider>/<model>.json, e.g. openai/gpt-4.1 | | Build metadata | api/v1/meta.json (generated_at, last_verified_at) | | Hugging Face | aicostbudget-ai/ai-api-pricing and the price history dataset | | JavaScript / TypeScript | npm install ai-api-pricing: prices, comparisons and token-cost estimates (docs) | | AI assistants (MCP) | npx -y ai-api-pricing-mcp: read-only tools for prices, monthly estimates, comparisons and price changes (setup) |

Documentation: dataset page · model schema · pricing contract · methodology · distribution projections

Code examples

Python

import json
import urllib.request

url = "https://aicostbudget.github.io/ai-api-pricing-data/api/v1/models/openai/gpt-4.1.json"
with urllib.request.urlopen(url) as response:
    model = json.load(response)

print(model["provider_id"], model["model_id"], model["pricing"]["input"])

JavaScript

const url = "https://aicostbudget.github.io/ai-api-pricing-data/api/v1/models/openai/gpt-4.1.json";

async function main() {
  const response = await fetch(url);
  const model = await response.json();

  console.log(model.provider_id, model.model_id, model.pricing.input);
}

main();

Interactive Tools

• Calculate your API cost
• Compare AI model prices
• Track AI model price changes
• Analyze retry cost
• Compare model-switch cost
• Estimate prompt caching savings
• Find the cheapest LLM API for a use case
• Compare AI coding plans (Claude Code, Codex, Cursor, Copilot)

Pricing Beyond Chat Tokens

Some APIs bill per character, per image or per page instead of per chat token. These website pages compare them:

• Embedding API pricing (per 1M tokens)
• TTS API pricing (per 1M characters)
• Image generation API pricing (per image)
• OCR API pricing (per 1,000 pages)

Explore Pricing by Provider

| Provider | Pricing page | Coverage | | --- | --- | --- | | OpenAI | [OpenAI API pricing](https:/