Balance—Sign in
Conversation settings
Saved with this chat.
Saves the system prompt and the Advanced settings in this browser.
Advanced
Lower is more focused, higher is more varied. Some models ignore it.
Caps the length of each reply. Empty uses the model default.
How long the model thinks before it answers. More effort uses more output tokens.
Download this chat
Chat with Claude, GPT, Grok and more
You're on a free model. Add $3 or more to use every model.Add balance
Incognito chat. Not saved in this browser: it is gone when you start a new chat, open another one or leave the page.
Enter to send, Shift+Enter for a new line.Return adds a new line.Models can be wrong. Check important answers.
Try a starter
One chat for 11 models. Pick a model per conversation, switch whenever you like, and see what each reply cost. Turn on Web search and the model looks things up and cites its sources. Replies are billed from your balance at the per-token rates on the models page.
What a reply costs
All 11 modelsUS dollars per 1M tokens. A typical short reply is a few hundred tokens.
| Model | Input | Output |
|---|---|---|
| Claude Opus 5.5 | $4.40 | $22 |
| GPT-6.1 Sol | $2.20 | $11 |
| Claude Sonnet 5.5 | $2.20 | $11 |
| DeepSeek V4 Flash | $0.24 | $0.96 |
- 01
Create an account or sign in. What you typed here is kept.
- 02
Add $3 or more to start, then pick a model. You pay per token.
- 03
Chats stay in this browser. The same balance pays for API calls.