Will my code fit in the context window?
Which models can read your code or text in one go, with room left for the answer.
Result
It fits in 35 of 42 models in one go, with 32K tokens left for the answer.
| Model | Context | Your text | Of the window | Result |
|---|---|---|---|---|
| GPT-5.5OpenAI | 1.05M | ~500K | 48% | Fits |
| GPT-5.6 SolOpenAI | 1.05M | ~500K | 48% | Fits |
| GPT-6 AstraOpenAI | 1.05M | ~500K | 48% | Fits |
| GPT-6 LunaOpenAI | 1.05M | ~500K | 48% | Fits |
| GPT-6 SolOpenAI | 1.05M | ~500K | 48% | Fits |
| GPT-6.1 SolOpenAI | 1.05M | ~500K | 48% | Fits |
| DeepSeek V4 ProDeepSeek | 1.05M | ~500K | 48% | Fits |
| DeepSeek V4.1 FlashDeepSeek | 1.05M | ~500K | 48% | Fits |
| Gemini 3.1 Pro PreviewGoogle | 1.05M | ~500K | 48% | Fits |
| Gemini 3.5 Flash-LiteGoogle | 1.05M | ~500K | 48% | Fits |
| Gemini 3.6 FlashGoogle | 1.05M | ~500K | 48% | Fits |
| Gemini 3.8 FlashGoogle | 1.05M | ~500K | 48% | Fits |
| Gemini 4 ArgonGoogle | 1.05M | ~500K | 48% | Fits |
| GLM-5.3Z.ai | 1.05M | ~500K | 48% | Fits |
| GLM-5.3 FlashZ.ai | 1.05M | ~500K | 48% | Fits |
| Kimi K3Moonshot AI | 1.05M | ~500K | 48% | Fits |
| MiMo-V2.6-ProXiaomi | 1.05M | ~500K | 48% | Fits |
| Mistral Large 4Mistral | 1.05M | ~500K | 48% | Fits |
| Muse Spark OpenMeta | 1.05M | ~500K | 48% | Fits |
| Muse WonderMeta | 1.05M | ~500K | 48% | Fits |
| Claude Fable 5Anthropic | 1M | ~650K | 65% | Fits |
| Claude Fable 5.1Anthropic | 1M | ~650K | 65% | Fits |
| Claude Haiku 5.5Anthropic | 1M | ~650K | 65% | Fits |
| Claude Opus 4.6Anthropic | 1M | ~500K | 50% | Fits |
| Claude Opus 4.7Anthropic | 1M | ~650K | 65% | Fits |
| Claude Opus 4.8Anthropic | 1M | ~650K | 65% | Fits |
| Claude Opus 5Anthropic | 1M | ~650K | 65% | Fits |
| Claude Opus 5.5Anthropic | 1M | ~650K | 65% | Fits |
| Claude Sonnet 4.6Anthropic | 1M | ~500K | 50% | Fits |
| Claude Sonnet 5Anthropic | 1M | ~650K | 65% | Fits |
| Claude Sonnet 5.5Anthropic | 1M | ~650K | 65% | Fits |
| Grok 4.3xAI | 1M | ~500K | 50% | Fits |
| MiniMax M3.1 FlashMiniMax | 1M | ~500K | 50% | Fits |
| Qwen3.8 27BAlibaba (Qwen) | 1M | ~500K | 50% | Fits |
| Qwen3.8 MaxAlibaba (Qwen) | 1M | ~500K | 50% | Fits |
| MiniMax M3MiniMax | 524K | ~500K | 95% | Too big |
| Grok 4.6xAI | 500K | ~500K | 100% | Too big |
| Grok 4.7xAI | 500K | ~500K | 100% | Too big |
| Kimi K2.6Moonshot AI | 262K | ~500K | 191% | Too big |
| Mistral Medium 3.5Mistral | 262K | ~500K | 191% | Too big |
| Claude Haiku 4.5Anthropic | 200K | ~500K | 250% | Too big |
| gpt-oss-120bOpenAI | 131K | ~500K | 381% | Too big |
How we work it out
We estimate tokens as characters ÷ 4, and add 30% for Claude models from 4.7 on, whose tokenizer counts more tokens for the same text[1]. A model fits when your text plus the room for the answer is within its context window.
Questions
How accurate is the token estimate?
It is a rule of thumb: about 4 characters per token for English text and code. Real counts depend on the tokenizer and the language, so leave some room.
Why do newer Claude models count more tokens?
Anthropic says Claude 4.7 and later models use a newer tokenizer that produces about 30% more tokens for the same text. The calculator adds that for those models.
Should I fill the whole window?
Rarely. Agents work better with a focused context, and every call resends it, so a full window is also the most expensive way to work.
More tools
- Compare AI modelsEvery pair, side by side
- AI token cost calculatorWhat your tokens cost on every model
- Prompt caching calculatorHow much prompt caching saves you
- Claude plan vs API calculatorSubscription or pay-as-you-go?
- Free limit reset trackerIs a free Claude or Codex reset live right now?
- Claude usage limit trackerDid the 5-hour or weekly limit go up or down?
- Claude Code usage dashboardYour tokens, models, cost and busiest hours, from one command
Sources
Numbers from the makers' own pages, checked by hand, last read . List prices in US dollars before tax. Spotted a mistake? office@shuriken.sh.
Made by Shuriken, the harness for coding agents. Its hot swapping moves your chats to another of your accounts or API keys when one nears its limit. Use only accounts you own and are entitled to use.
Try Shuriken on your own computerYou're signed in
Add your Claude, Codex and Gemini accounts once. Shuriken switches between them so your agents keep working.Your accounts, projects and chats are in the app.
Coming soon