01
Switch models mid-project
Every model is grouped by provider and series in one picker, with context length and live per-million token pricing next to each name.
New5 credits for free after you make an accountOne workspace for
One workspace for
every leading AI model
Claude, GPT, Gemini, Llama, Grok and hundreds more behind a single login and a single balance. Pick the right model for the task instead of the one your subscription happens to cover.
- No subscription
- No card to start
- $1.00 per credit after that
Claude Sonnet 4.5
Turn these notes into a concise launch plan with clear owners.
Here’s a focused three-stage plan:
- 1. Positioning — Finalize the promise and proof points.
- 2. Production — Ship the core launch assets.
- 3. Distribution — Sequence owned and partner channels.
0.04 credits used
Message any model…
Model catalog
Every major provider, one bill
The catalog tracks OpenRouter, so model availability and pricing stay current without anything to configure.
OpenAI
GPT-5, GPT-4o, o-series
Anthropic
Claude Opus, Sonnet, Haiku
Gemini 2.5 Pro and Flash
Meta
Llama 4 and Llama 3.x
xAI
Grok 4 and Grok Code
DeepSeek
DeepSeek V3 and R1
Mistral
Mistral Large and Small
Qwen
Qwen3 and Qwen Coder
What you get
A workspace, not a demo
The details that decide whether a chat tool is usable on a Tuesday afternoon.
02
Answers that survive a reload
Responses are generated and saved server-side. Close the tab mid-answer, come back, and the conversation picks up where it left off.
03
Built for real output
Markdown, tables, LaTeX, and syntax-highlighted code with one-click copy — so answers are usable without reformatting them first.
04
Your history, your account
Conversations are scoped to your account behind Clerk authentication, and can be renamed or deleted at any time.
How it works
Three steps to a first answer
Step 1
Create an account
Sign up with email or a social login. 5 credits land in your balance once your email is verified.
Step 2
Pick a model
Search the catalog by provider, series, or name. Your last choice is remembered per conversation.
Step 3
Pay for what you use
Each response deducts credits based on the model and the tokens it actually processed. Nothing recurring.
Pricing
Pay for responses, not seats
One balance covers every model. Credits do not expire and there is nothing to cancel.
Free
5credits on signup
Enough to try several models on real work before spending anything.
- Full model catalog
- Saved conversation history
- No payment details required
Pay as you go
$1.00per credit
One credit covers $0.70 of model usage at provider rates. Buy 5 or 200 — the price per credit is the same.
- Charged on the provider's reported cost
- Balance updates with every response
- No monthly minimum or seat count
FAQ
Questions, answered
Which models can I use?
Anything in the live OpenRouter catalog — several hundred models from OpenAI, Anthropic, Google, Meta, xAI, DeepSeek, Mistral, Qwen and more. New releases appear as soon as OpenRouter lists them.
How exactly are credits charged?
Each response is billed on the provider's reported cost for that request. One credit covers $0.70 of model usage, so a short exchange with a small model costs a fraction of a credit and a long one with a frontier model costs more. Your balance updates as soon as the response settles.
Do I need a subscription?
No. There is no monthly fee and no minimum. You start with 5 free credits and buy more only when you want to, at $1.00 per credit.
Are there usage limits?
There are fair-use guardrails: one response generating at a time per account, a cap on requests per hour, and a per-request spend ceiling. They exist to prevent runaway costs, not to ration normal use.
Who can see my conversations?
Conversations are tied to your authenticated account and are not shared between users. Prompts are sent to the provider you select in order to generate a response.
Start with the model that fits the task
Create an account, get 5 credits, and keep every conversation in one place.
Create your account