I’m having a problem with the multi-model response feature in TypingMind. At the moment, it’s not possible to save multiple model responses permanently. Please change this. I suggest that, in future, all model responses should be saved permanently in the chat.

Please consider adding Opencode Go and Opencode Zen as a provider. I extensively use these on other platforms and would be nice to have it on typingmind too.

Currently, selecting a prompt from the library automatically overwrites all existing text in the chat input box. Ideally, the default behavior should be updated to simply append the selected prompt, but if that isn't feasible, please add an 'Append to current input' checkbox within the prompt editor. Great work on closing the previous ticket ('Prompt Library Browsing w/o Erasing Prompt') a year ago, which successfully fixed text being erased when simply opening the library! This new request focuses strictly on the replacement that happens when inserting the prompt, giving users a seamless way to combine custom text with saved templates.

Without this feature it is very difficult to see longs chants because in typing mind chat search options We can see the chat content but by clicking,We cannot Jump to that particular question in the chat. This feature ChatGPT has developed and this is very good feature take it seriously.

In fact, I have Claude’s new Cowork feature in mind.

Setapp now supports GPT-5.4, GPT-5.4-Pro, GPT-5.4-mini, GPT-5.5, Claude Sonnet 4.6, Gemini 3.1 Pro, and possibly others as I have seen these supported in some other Setapp AI apps. Can you add these to TypingMind? Thanks!

I suggest TypingMind add a feature to allow dynamic context to fetch/lookup data directly from the Knowledge Base, so that AI agents and chat prompts can access and use up-to-date internal documents more conveniently.

I think in the future you could add optionally some local models to make some part of the work locally, in example audio to text, and save; or even some simple reasoning tasks. :)

I’d like to suggest a feature that could make TypingMind even better for users who run local LLMs. While TypingMind already supports custom OpenAI-compatible endpoints, configuring local inference servers such as Ollama, LM Studio, LocalAI, vLLM, llama.cpp, and similar solutions still requires manual setup. It would be fantastic to have a dedicated Local Models section that makes connecting to local LLMs much easier and more user-friendly. Some ideas that could improve the experience include: Automatic detection of local inference servers running on localhost (especially Ollama and LM Studio). One-click connection to supported local backends. Built-in presets for Ollama, LM Studio, LocalAI, vLLM, llama.cpp server, and other OpenAI-compatible APIs. Automatic discovery and listing of available models instead of requiring users to manually enter model names. A connection test with clear diagnostics when a server isn’t reachable. An optional setup wizard to guide new users through the configuration process. A streamlined setup process would significantly improve the user experience while making TypingMind an even more compelling choice for anyone who wants to combine cloud models with locally hosted models. Thank you for taking the time to read this suggestion and for continuously improving such an excellent product. I hope you’ll consider adding these enhancements in a future release.

Current login solutions involve one time passcode or passkey. Both solutions require external dependencies such as access to email and access to password manager. QR login allows typing mind to be used on public computer without the risk of password being stolen. It also requires less mental effort to use compared to one time passcode.

I'd like to see a table view on the provider model name screen that shows model details (release date, pricing, etc) with the ability to sort by any column. To see this information today, I have to select each model individually. To accommodate that view, I'd redo this block. It takes up way too much space currently: I'd suggest you move this cntent block into a 'Model Details' tab and put it next to the models | api keys | global settings navigation above the model list, and use this space for a sortable table. Or just shift it further over to the right. it's using up a lot of space currently. It's useful, but uses more space then necessary. writingmate.ai does a really good job at this; see screenshot below. This is what I’d like to see when I select a model provider; a table-based layout of all the models therein along w/ relevant information on each LLM the provider offers: I can sort by model, author, date dated, context, input and output, and i can enable/disable models directly from this screen. Since so many models are added each week, giving me the flexibility to sort > enable/disbable based on needs makes it faster, quicker and easier to choose a model based on credit usage. Sometimes I want the cheapest model, sometimes middle of the pack, sometimes high-tiered ones. but right now, I have to choose the provider, select the model, look in that right pane to see token cost, etc. Maybe this table view is where the 'more model details' shows up? add a 'view details' link next to the model name? If i select that, a modal slides in from the right with the model overview/parameters shown? This would streamline things greatly!

Hi Tony, Thank you for the update on expanded model provider support and response API endpoint support. I am very happy to see my long-standing request has been fulfilled. Please have option to display Credit balance under api key management, many provider have endpoint to get credit balance details Like for vercel: Base URL: https://ai-gateway.vercel.sh/v1 Endpoint: GET /credits Json Key: balance. for openrouter: curl https://openrouter.ai/api/v1/credits -H "Authorization: Bearer " { "data": { "total_credits": 100.5, "total_usage": 25.75 } } Thank you for the continued improvements to TypingMind.

Is it possible to share chats across users so that multiple users can speak into a conversation?

We have built agents for our team members, and having admin control over these showing up granularly would simplify the experience for our team. Since we are using agents, a starting model is not relevant (our system runs only on agents). Additionally having it optional from our end for our users to be able to append or not would be useful. All of the features in this section are great — they just need to be separated out so we can control our system better.

It would be helpful for agents to be able to create inline buttons during chat when the agent is presenting options. For example “Would you like me to A, B, or C” and the buttons appear clickable in the chat (similar to the conversation starter buttons). Perplexity and others offer this functionality. It’s helpful both on desktop and mobile!

Hi TypingMind Team, I would love to request native Korean language support for the platform. South Korea has an exceptionally high and rapidly growing adoption rate for AI tools, with a massive community of creators, developers, and professionals who use AI daily. Introducing Korean localization would break language barriers for many potential users and significantly boost TypingMind's growth and user base in the Korean market. I would be more than happy to provide active feedback to help with the localization process if needed. Please consider adding Korean to the roadmap! Thank you for building such an amazing tool.

I would like TypingMind to support connecting to Ollama Cloud using its official OpenAI-compatible API endpoint (https://ollama.com/v1/). Ollama Cloud offers hosted versions of models like Llama, Gemma, Nemotron, but currently TypingMind’s proxy does not support this endpoint, and browser CORS blocks direct access. Please add proxy and/or native support for Ollama Cloud’s API endpoint, so that users can use Ollama Cloud models within TypingMind just like OpenAI or Mistral. This would enable access to the latest open-source models with easy paid hosting. Thank you!

I hope there is a button to allow Multi-agent Debate Mode. This is I think even more useful than parallel agent output. MAD can save so much time analyzing a complicated topic, especially those with many sub-topics (ex: should universal basic income be implemented in the next years?) [MECHANICS] - Two or more AI language models debate a user-provided topic through structured, round-based interaction. - The human user: a) observes all outputs at every stage; b) may intervene between rounds; c) remains the final arbiter of whether consensus has been reached. [FEATURE REQUEST] - Please add a file attachment tool that allows the human user to upload and provide multiple reference files simultaneously to all participating LLMs. - The file attachment tool should support, at minimum: 1. PDF files 2. .txt files 3. Markdown files 4. docx/MS Office/LibreOffice files - The attached files should be available as shared reference material for any/all LLMs during the debate. - The human user should be able to attach files before the debate begins and, ideally, also between rounds when asking follow-up questions or requesting clarification. ** A strong multi-LLM debate protocol should not imitate a loose human panel discussion. It should behave more like a formal academic debate with controlled rounds, visible transcripts, forced engagement. - A multi-agent debate protocol should also offer the option to: - have the MAD protocol decide whether there has been full consensus on the entire list of topics, or - allow a human judge to decides whether true consensus has actually been reached on any or all of the discussion points.

Quick scroll to top and scroll to bottom of chat buttons Quick scroll to up or down pinned items - also iamges, etc. Shift select for multi select in chat window (for delete, etc) rather than just one by one button to delete all images included in chat thus far [ this is critical when changing models to a non-image model within an existing chat with images ] Allow ordering of model for priority in the drop-down. Can’t see full model names in the prompt model select area - really annoying to have to hover to see the full model version etc. Allow adjusting width of chats - even though the window is wide, the chat only reaches a narrow width. Why isn’t this adjustable?