AI Model Context Window Reference
A searchable, filterable reference table of context window sizes, pricing, and key specs for every major LLM — GPT-4, Claude, Gemini, Llama, Mistral, and more. Updated May 2026.
| Model | Provider | Context | In $/1M | Out $/1M | Notes |
|---|
Data based on publicly published specifications and pricing as of May 2026. Pricing, context windows, and model availability change frequently — verify with the provider before production use.
Disclaimer: Free tool provided “as is” by MonitorGiant. No warranty or liability for any data loss, security issues, or infrastructure problems arising from use of this tool. Results are for informational purposes only. · A Free Tool by MonitorGiant
What is AI Model Context Window Reference?
The context window is the amount of text an LLM can process in a single call — including both the prompt (input) and the response (output). Larger context windows allow you to send entire codebases, long legal documents, or multi-turn conversation histories without chunking. As of mid-2025, context windows range from 16k tokens (GPT-3.5 Turbo) to 1M+ tokens (Gemini 2.0 Flash, GPT-4.1). A larger window does not always mean better performance — models often exhibit 'lost in the middle' behaviour where information in the middle of a very long context receives less attention than content at the start or end.
How to use this tool
- 1 Use the search box to filter by model name, provider, or keyword. Use the Provider dropdown to compare models from a single company.
- 2 Use the Context Size filter to show only models above a minimum token threshold — useful when you need a specific window size for your use case.
- 3 Click any column header to sort. Sort by Context to find the largest windows, by input price for cheapest models, or alphabetically by name.
- 4 Read the colour-coded context size badges: purple = 1M+ tokens, blue = 200k+, cyan = 128k+, teal = 32k+.
When would you use this?
- Developers selecting a model for a new feature who want to compare context windows and pricing in one place.
- Teams migrating from one provider to another checking that existing prompts fit in the target model's context.
- Prompt engineers identifying which models support the 1M-token context window needed for full-document analysis.
- Budget-conscious teams sorting by input price to find the cheapest model that meets their context size requirement.
How works
- 1
Search or filter
Type any model name, provider, or feature keyword into the search box. Use the Provider filter to compare models from one company, or the Context Size filter to show only models above a token threshold.
- 2
Sort any column
Click a column header to sort. Click again to reverse. Sort by Context to find the largest windows, by input price to find the cheapest models, or alphabetically by model name.
- 3
Read the context size colour codes
Purple = 1M+ tokens, Blue = 200k+, Cyan = 128k+, Teal = 32k+. This lets you quickly identify the context tier at a glance.
- 4
Check open-weight models
Models with open = 0 pricing are open-weight (e.g. Llama 3, Mixtral). They are free to run yourself but have inference costs if used via a hosted API such as Groq, Together, or Fireworks.
This is a fully static reference page. No API calls are made. All data is hardcoded from publicly published sources as of May 2026.
Comments & Feedback
Found a bug? Have a suggestion? We'd love to hear from you.
Related Tools
From the makers of this tool
Need deeper observability?
MonitorGiant tracks real-time AI performance, infrastructure health, and system reliability — far beyond what free utilities can show.