Compare 49+ LLM models side by side. Filter by use case, budget, or provider to find the best model for your project. Sort by price, context window, or provider.
| Model | Provider | Input $/M ▼ | Output $/M | Context | Best For |
|---|
Found the right model? Estimate your costs or generate the API code.
Tokonomics works with every model listed here. One URL change and every call is metered with budget alerts.
Start Free →Start with your use case. For customer-facing chatbots, prioritize quality (GPT-4o, Claude Sonnet). For high-volume tasks like classification or extraction, use a cheap model (GPT-4o-mini, DeepSeek Chat). For coding, Sonnet and GPT-4o lead benchmarks. Use the filters above to narrow down by budget and use case.
The context window is the maximum number of tokens a model can process in a single request (input + output combined). A 128K context window means roughly 96,000 words. Larger contexts let you process long documents but cost more per call. For RAG use cases, 128K+ is ideal.
Smaller, distilled models (GPT-4o-mini, Haiku, DeepSeek Chat) trade some reasoning ability for 10-50x lower cost. For 80% of production use cases — classification, extraction, simple Q&A — they perform identically to premium models. See our cheapest LLM guide for benchmarks.
We sync pricing from official provider documentation weekly. Last update includes all major providers: OpenAI, Anthropic, Google, DeepSeek, Mistral, Groq, xAI, and Cohere. Use our cost calculator for detailed monthly cost projections.
The budget-first AI cost metering proxy for any stack. Track every LLM token, set budget alerts, and never get surprised by your AI bill again.