AI Model Pricing Comparator

Compare LLM pricing across 15+ providers in real-time. Find the cheapest model for your use case in seconds.

What it does

Aggregates live pricing data from 380+ AI models across major providers:

• OpenRouter (340+ models) — OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, Cohere, xAI, and more
• Venice AI (43 models) — Uncensored models, unique offerings

All data is fetched in real-time from provider APIs (no stale caches). Results are normalized to a common schema with USD per million tokens pricing.

Use cases

• "What's the cheapest model with vision?" — Filter by modality, sort by price
• "Compare all Claude models" — Search by name, see every provider's price
• "Models under $1/MTok with 128K+ context" — Filter by price and context window
• "What does GPT-4o cost on different providers?" — Cross-provider price comparison
• Cost optimization for AI applications — Find cheaper alternatives to expensive models

Input

Parameter	Type	Description	Default
query	string	Search model names and descriptions	(all)
maxInputPrice	number	Max input price in $/MTok	(no limit)
maxOutputPrice	number	Max output price in $/MTok	(no limit)
minContextWindow	integer	Min context window in tokens	0
modality	string	Filter: any, text, multimodal, image-gen, audio	any
capabilities	string[]	Required: vision, function-calling, reasoning, web-search	(none)
providers	string[]	Limit to specific providers	(all)
sortBy	string	input_price, output_price, context_window, name	input_price
includeFree	boolean	Include $0 (free tier) models	true
maxResults	integer	Max models to return (0 = all)	0

Output

Each model in the dataset includes:

{
    "name": "Anthropic: Claude Sonnet 4.6",
    "model_id": "anthropic/claude-sonnet-4.6",
    "provider": "Anthropic",
    "source": "openrouter",
    "input_price_per_mtok": 3.0,
    "output_price_per_mtok": 15.0,
    "context_window": 1000000,
    "max_output_tokens": 64000,
    "modality": "multimodal",
    "input_modalities": ["text", "image"],
    "output_modalities": ["text"],
    "capabilities": ["vision"],
    "description": "...",
    "huggingface_id": "...",
    "fetched_at": "2026-03-11T16:45:00+00:00"
}

Examples

Cheapest non-free models:

{"includeFree": false, "sortBy": "input_price", "maxResults": 20}

All Claude models compared:

{"query": "claude", "sortBy": "input_price"}

Cheapest multimodal (vision) models:

{"modality": "multimodal", "includeFree": false, "sortBy": "input_price", "maxResults": 10}

Models under $1/MTok with 100K+ context:

{"maxInputPrice": 1.0, "minContextWindow": 100000, "sortBy": "input_price"}

How it works

1. Fetches model data concurrently from OpenRouter and Venice AI public APIs
2. Normalizes pricing to USD per million tokens
3. Classifies modality (text, multimodal, image-gen, audio)
4. Extracts capabilities (vision, function-calling, reasoning, web-search)
5. Applies your filters and sorting
6. Pushes results to the Apify dataset

No API keys required for the default providers. Runs in under 5 seconds.