> ## Documentation Index > Fetch the complete documentation index at: https://docs.enconvo.ai/llms.txt > Use this file to discover all available pages before exploring further. # Phone Chats > Chat with AI models on the iPhone itself: pick the default model, add providers and keys, or sign in with ChatGPT, Grok, Le Chat, GitHub Copilot, or Antigravity. Phone chats are the iPhone app's own chats. They run on the phone, not on your Mac, so they work without a paired Mac. Each Phone chat uses a model through Enconvo Cloud points, your own API key, or a plan you sign in to on the iPhone. Phone chats and personal instructions sync through your Enconvo account. To continue a chat that runs on your Mac instead, see [iPhone App](https://docs.enconvo.ai/integrations/iphone-app.md). ## Choose the default model 1. On the iPhone, open **Settings → AI Model**. 2. Tap **Default model** and pick a model. New chats start with it. 3. To change one chat's model, use the model picker in that chat. Enconvo on your Mac starts with the same model when the Mac has that provider. If the Mac's model isn't available on the iPhone, the page says so under **Default model**, and suggests what to add or sign in to. ## Add a key or a provider **AI Model** lists Enconvo Cloud and every provider you can use on the phone. 1. Tap a provider, then **Add an API key**, and paste a key you created on the provider's site. 2. To add a provider that isn't in the list, tap **Add a provider** and pick a provider, **OpenAI compatible**, or **Anthropic compatible**. A compatible provider needs the address of a server that speaks OpenAI's Chat Completions or Responses API, or Anthropic's Messages API. 3. If a provider doesn't list a model you need, tap **Add a model by ID** on its page. Keys stay in your iCloud Keychain and are sent only to the provider they belong to. ## Sign in with a plan Five providers let Phone chats use a plan you already pay for, with no API key and no Cloud points: | Provider | Plan | Button | | - | - | - | | OpenAI | ChatGPT | **Sign in with ChatGPT** | | xAI | Grok | **Sign in with Grok** | | Mistral | Le Chat | **Sign in with Mistral** | | GitHub Copilot | Copilot | **Sign in with GitHub** | | Antigravity | Antigravity (Google) | **Sign in with Google** | **1. Open the provider** In **Settings → AI Model**, tap the provider. For OpenAI, xAI, and Mistral, choose the plan's card under **Authentication**. The other card is **API Key**, and **In use** marks the one Phone chats use now. **2. Sign in** Tap the sign-in button and finish signing in on the page that opens. Use the account that has the plan. **3. Check the result** Back on the provider's page, you see your plan, your account, and how many models are on your plan. Pick a model, or tap **Set as Default** to start new chats with it. A plan's sign-in stays on this iPhone. It doesn't sync to iCloud or to your Mac, and the Mac keeps its own sign-in. The plan's own usage limits apply. ### GitHub Copilot After **Sign in with GitHub**, the app shows a GitHub code. **Continue** copies the code and opens GitHub's page, where you paste it and authorize Enconvo. The code stays on screen while GitHub's page is open, so you can copy it again. The model list shows only the models your Copilot plan can use. Copilot Free offers fewer models than the paid plans. ### Antigravity Before each sign-in, the app shows **Before you sign in to Antigravity**. Enconvo signs in with the Antigravity app's own Google client, and Google may suspend an Antigravity account it sees used outside Antigravity. Use an account you can afford to lose. **Agree** goes on to Google's sign-in, and **Cancel** leaves everything as it was. Antigravity offers the Gemini and Claude models on your plan, each with the reasoning effort levels Antigravity has for it. ### ChatGPT usage OpenAI's page shows how much of your ChatGPT plan's 5-hour and weekly limits is left, when each limit resets, and when the plan renews. Tap **Refresh usage** in the account menu to update it. When a limit runs out, the page says so until it resets. ### Sign out Open the provider and choose **Sign out** in the account menu. Chats that use the plan need another model after you sign out. For Mistral, the key the sign-in made stays in your Mistral workspace. ## Reasoning effort When you pick a model that thinks, the picker goes on to its reasoning efforts. The effort you pick for a model shows beside its provider in the chat's header and applies to every chat on that model. A plan's page lists only the levels your plan offers. ## Long chats When a long chat nears its model's context window, the same model summarizes the older messages, as on the Mac. The chat keeps every message, and a **Context compacted** mark shows where the summary starts. This is always on. ## Troubleshooting | What you see | What to do | | - | - | | **ChatGPT sign-in didn't finish** (or Grok, Mistral, GitHub Copilot, Antigravity) | Tap **Sign in again** and finish signing in on the page that opens. | | **Your plan doesn't offer** a model | Choose another model from the plan's list. On Copilot Free, only some models are offered. | | **Listing models…** doesn't finish | Check the iPhone's network connection, then leave the provider's page and open it again. | | The Mac's default model isn't on the iPhone | Add the provider's key, or sign in to its plan, as the note under **Default model** suggests. | ## Next steps > **[iPhone App](https://docs.enconvo.ai/integrations/iphone-app.md)** > > Pair the iPhone with your Mac and continue its chats. > **[ChatGPT Subscription](https://docs.enconvo.ai/providers/openai_subscription.md)** > > The ChatGPT sign-in on the Mac. > **[GitHub Copilot](https://docs.enconvo.ai/providers/github-copilot.md)** > > The Copilot sign-in on the Mac. > **[Antigravity](https://docs.enconvo.ai/providers/antigravity.md)** > > The Antigravity sign-in on the Mac. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.