[v2.0.2] What's New • New models arrive faster. SayFlow now keeps its model list up to date automatically, so new models from OpenAI, Anthropic, Google, SpaceXSI, and DeepSeek can show up without waiting for an app update. • Support Anthropic 'Claude Haiku 5.5'. Claude Haiku 4.5 is now under Others. • Support Google 'Gemini 4 Argon' and 'Nano Banana 2.1'. • Support OpenAI 'GPT-6.1 Sol'. • Translation from other apps now uses the same reasoning and temperature settings as your chats. • Removed the custom model name field. Choose models from the list instead. • Sending the first message in a new chat now animates smoothly, and the keyboard stays open. • Fixed an error that could stop Claude from replying after it used web search or skills. • Grok searches the web and X faster and uses far fewer tokens per question. • Error messages from Grok now explain what went wrong, like running out of credits, instead of showing only an error code. • A fresh look for chat bubbles • The reasoning bubble is now a round, puffy thought cloud, just like a comic. • Stability improvements and bug fixes.

[v2.0.1] What's New • Added a Local Model provider that runs AI models entirely on the device, with no API key and no data sent to a server. • Merged Apple Intelligence into Local Model. Choose Apple Intelligence, Gemma 4 E2B, Qwen3.5 0.8B, Qwen3.5 2B, or your own GGUF model. • Apple Intelligence is the default on supported devices. Gemma 4 E2B is the default on other devices. • Models are offered based on the device's available memory (RAM) instead of the device model. • Added a Manage Models screen: - Download a model with the download icon, then watch its progress and cancel if needed. - Delete a downloaded model with the trash icon, after a confirmation prompt. - Each model shows a one-line description, its size, and its memory status. • Downloads continue in the background, even if the app is closed. • Import your own GGUF models from the Files app. SayFlow checks the file format, chat format, model architecture, and memory before importing. • Added image input for Gemma 4. After downloading Image Input (557 MB), you can attach photos to local chats. • Local models are available in Settings and in the chat model menu, with model logos. Each chat remembers its model. • Updated the provider icon and related settings text. Chat • Sent messages now rise from the composer into place. • The keyboard now stays open after sending a message. • Bold and italic text now render correctly next to inline math. • The "Writing a reply…" bar no longer covers the composer while a chat or translation screen is open. Fixes • Fixed the control widget label. • Fixed an issue where a chat could suddenly turn blank right after opening it. • Scrolling in conversations is steadier, with fewer unexpected jumps. • Sent messages now glide into place with a smoother animation. • Tap anywhere in the conversation to dismiss the keyboard. • Automatic chat compression now waits until a conversation is actually long, so short chats are no longer summarized after just a few messages. • Compressed summaries now build on the previous summary, making repeat compressions faster and more accurate. • The keyboard now stays open after sending, so you can keep typing.

[v2.0.0] What's New A Brand-New Layout • SayFlow now uses a tab bar. Chats, Translation, and Archives each have their own tab. • Pinned chats appear at the top of the Chats tab, and your other chats are listed below under Recents. • Search from the bar at the top of the Chats tab. Results include your chats, archived chats, and translation history, and tapping a result takes you straight to the matching message. • Start a new chat with your default provider using the new + button, or tap the provider button next to it to choose a different one. • While a reply is being written or a translation is in progress, a small bar appears above the tab bar. Tap it to jump back in. (iOS 26.1 or later) Chat List • Swipe left on a chat to pin, archive, or delete it. (iOS 27 or later) • Deleting, archiving, and pinning chats now animate smoothly. • In an archive, Select mode hides the tab bar so the Delete button is easy to reach, and selected chats are shown more clearly. • Separator lines in lists now line up with the text. Translation • Start a new translation with the + button in the Translation tab. New Models • OpenAI: Added GPT-6 Sol and GPT-6 Luna. GPT-5.6 Sol, Terra, and Luna are now grouped under GPT-5.6. • GPT Image 2.5: Added Flare and Sunburst, with new Extra High and Max quality options. • Anthropic: Added Claude Opus 5.5, Sonnet 5.5. Opus 5 and Sonnet 5 have moved to Others. Fixes • Fixed an issue where the provider menu could show an oversized logo. • Long model names are now shortened so the buttons in the chat toolbar stay visible.

[v1.0.9] * Added an onboarding screen shown on the first launch after installation. * Removed cost estimation tracking from estimated token information. * Support OpenAI ‘GPT-6 Astra’ API. * Support Google ‘Gemini 3.7 Flash’, ‘Gemini 3.8 Flash’ API. * Bug fixes and improvements.

[v1.0.8] * Support Google ‘Gemini 3.6 Flash’, ‘Gemini 3.5 Flash Lite’ API. * Support Anthropic ‘Opus5’ API. * Support SpaceXAI ‘Grok 4.6’ API. * Refresh New DeepSeek V4 Pro, Flash API. * Remove API pricing info. (redirect to provider docs instead) * Other bug fixes.

[v1.0.7] * Moved Gemini image models to the latest stable APIs and improved continuous image editing and model-specific aspect ratios. * Gemini Google Search responses now include sources and Search Suggestions. * The Translation Extension now follows reasoning effort, Thinking, and temperature settings more accurately. * Improved reasoning controls and automatic title generation for Grok 4.20 Multi-Agent. * Improved voice input reliability and privacy disclosures. * Fixed issues affecting provider requests, token metrics, caching, and conversation stability.

-Some UI Improvements. -Added a reminder that AI can make mistakes, so important information should be cross-checked. -AI Provider Name and Logo Change - Existing 'xAI' Changed to 'SpaceXAI’. -Support OpenAI 'GPT-5.6' series API. -Support Grok Imagine - Image Editing. -Support SpaceXAI ‘Grok 4.5’ API. -Support Google ‘Nano Banana Lite’ API. -Support Anthropic ‘Sonnet 5’ API. -Moved default provider selection to its own page, with options to choose which providers are shown and reorder their display order. -You can now save translation results as SayFlow conversations. From the translation screen, archive a result or open it directly as a new chat session. -Improved the reliability of conversation saving and recovery. -Made stopping responses and switching or deleting chats more accurate. -Improved the safety and reliability of attachments, shared photos, and imported translations. -Improved streaming responses and provider-specific error handling. -Improved performance for long conversations and image exports. -Enhanced accessibility and overall app stability.

[v1.0.4] * Added support for Anthropic’s Claude Opus 4.8 API * App optimizations

[v1.0.3] * Improved some UI elements * Optimizations related to reasoning process output * Fixed an issue where built-in wallpapers, except for “None,” were affected by Dark Mode * Fixed an issue where the temperature adjustment button was shown for the Gemini 3.5 Flash model * Fixed an issue where copy, share, regenerate, and response time options were displayed in the reasoning process * Fixed an issue where additional images could not be generated in the same session after generating an image with Nano Banana-series models * Fixed an issue where image generation models could occasionally output two copies of the same image * Updated the cost information for the DeepSeek V4 Pro model * Added a share button to the photo preview screen * Added pinch-to-zoom in/out support on the photo preview screen

[v1.0.2] * Support Gemini 3.5 Flash API * Modified Anthropic so that when Thinking is not disabled, the temperature is forcibly fixed at 1.0 * Added haptic feedback and toast notifications to some button interactions * Added external web search models (Tavily, Serper) * Added the ability to choose between the built-in web model or an external web model when invoking the web search tool * Increased the size of user and AI profiles * Added support for attaching images by pasting them into the message input field * Added an option to display shadows on chat bubbles * Modified reasoning output to be shown in real time even when responses are not streamed * Completely redesigned the message input UI, button layout, temperature panel, and Anthropic Thinking panel structure * Replaced the additional actions button in the message input field from the old panel style to a sheet style * Improved the layout of the prompt popover in the additional actions sheet * Disabled the send button while conversation compression is in progress and added a toast notification * Added SKILL.md support * Added Grok Imagine API support * Updated model codes for Google Gemini series models * Fixed incorrect labeling of Gemini 3 Flash as 3.1 * Added non-reasoning environment support for the Grok 4.3 API * Removed the Grok 4.1 API due to end of support * Added support for applying colors to archive icons * Added 3 new wallpapers (Glow) * Added a new app icon (Dot Fairy) * Added the ability to change icons and colors when editing generated archives * Fixed an issue where selecting the “Star Fairy” icon applied a different icon than shown in the preview * Improved the design of the sidebar Pinned and Archive layouts * Improved the way conversation menus are displayed when long-pressing chats in the sidebar to make the selected conversation clearer * Reorganized settings into categorized sections for improved accessibility * Fixed lock screen widget display issues * Fixed chat room metadata text to use contrast-aware colors based on the background * Fixed an issue where the profile button disappeared when the profile name exceeded a certain length * Fixed various minor bugs

[v1.0.1] - UI Improvements - Added an expand button when the chat input field exceeds two lines, allowing users to write in a larger sheet - Added the ability to change the entire conversation background by selecting a photo from the photo library - Fixed an issue where scrolling could jump in long conversations - Added support for using custom prompts for translation - Fixed an issue where the Lock Screen shortcut widget did not work - Fixed an issue where chat bubbles with attached photos could overlap the timestamp area