Short, honest answers about what UsingOpen is, what it costs, how availability works, and what happens to your conversations. If something is missing, ask on X.
Frequently asked questions
What is UsingOpen?
UsingOpen is a free chat front end for open-weight AI models. You pick a model from the catalog, send a prompt, and the reply streams back. The model runs upstream; the site keeps the API key off your device and shows live availability, context size and price for each model.
Is UsingOpen free?
Yes. There is no account and no paid tier. Each user gets 10 free prompts per day, reset at midnight UTC, and no card is required. Every live model is included; a model paused for capacity is unavailable until it returns. The per-token prices on model pages are upstream API rates, not charges on this site.
Do I need an account or email address?
No. UsingOpen does not ask for an email, phone number or password. Chat history is stored in your browser, so there is no account to sign back into.
What is the age requirement?
UsingOpen has no account system, so it does not ask for your age and there is no age gate. That does not remove the rules where you live: you are responsible for meeting any minimum age that applies to AI services in your jurisdiction, and minors should use the site with a parent or guardian's permission. Model answers can be wrong or unsuitable, so treat them as a starting point rather than expert advice.
Does UsingOpen actually work?
Yes — the chat is live, but no free service can guarantee every model at every moment. Availability depends on upstream capacity: the picker shows live status, paused models are greyed out as "Adding capacity", and the rest stay usable. If a model fails or is paused, try another or return later. A request that fails before it reaches a model does not use up a daily prompt.
Why do some models say "Adding capacity"?
The upstream provider pauses models while it brings more capacity online. UsingOpen mirrors that state instead of offering a model that would immediately error. A paused model is not gone: it returns to the picker when capacity is available, and other models remain selectable in the meantime.
Which models are available?
The catalog spans text, coding, reasoning and multimodal models, including DeepSeek V4, Qwen3.8, GLM 5.3, Gemma 4, GPT-OSS, Kimi K3 and MiniMax M3. The full list with context windows, prices and capabilities is on the models page; the picker on the chat page shows what is ready right now.
Is my chat history private?
Your history is saved in your browser (IndexedDB) and is never stored on UsingOpen's servers. To generate a reply, the active conversation is sent through the backend to the upstream provider serving the model. Each model shows a retention hint: zero retention where the provider offers it, otherwise assume normal provider logging. For quota, the backend uses a client ID and your IP address. Do not paste secrets into any AI chat.
Can I use UsingOpen on my phone?
Yes. The page is responsive and installable as a standalone app from the browser menu. It is tuned for small screens, the on-screen keyboard and light/dark themes, and history stays on the device.
How is UsingOpen different from ChatGPT?
UsingOpen is a model picker for open-weight models rather than one assistant. You choose the model per conversation, can compare two side by side, and never create an account. It is narrower by design: no image generation, voice mode or account sync. Use a mainstream chatbot when you need those features.
How do I report a problem or give feedback?
Send a message to @UsingOpen on X. Include the model name and what happened — a failed prompt, a wrong status label or a layout issue — so it can be reproduced.
Start a free chat · Browse the model catalog · Compare two models