Feature Requests

Please provide features like Create charts, Table, Diagrams, and Mindmap in response output upon request. Its ultimate feature that need everyone

Could you please add Claude Opus 4 and Claude Sonnet 4 to the available models on Merlin? These are the most recent and powerful releases from Anthropic, and having access to them would be incredibly valuable for tasks that require deep reasoning, creative writing, or detailed analysis. They would be a great addition to the platform’s lineup.

I think it would be awesome to have a "Prompt Enhancer" feature. It could use a clickable icon to take the user's prompt and refine it, which I believe would really improve LLM results and overall user experience. Using a more affordable model like Gemini 1.5 Flash or GPT 4o-mini could make this feasible and super useful.

It will be great to include open ai gpt-image-1 model for image generation

Currently the name of the chat starts with the original prompt, but that is not very specific and many times the prompts start the same way so its hard to determine what the chat was about when going through the history. I currently use this prompt to rename the chat, though its a very tedious and manual process: 'please give me a TLDR 5-7 words describing our chat here so I can use it as a title for renaming this chat' Maybe you can just integrate something like this that automatically renames the chat ever 4-5 prompts?

Google just released Gemini 2.5 Pro Experimental 03-25, and it would be great to have it available in Merlin! Given its improvements in reasoning and efficiency, it could be a strong option alongside the other top models. Would it be possible to integrate it soon so users can test its capabilities within Merlin?

Please help to add the feature which can allow to delete Image created by Bonkers by myself.

Enable image uploads for projects

This model seems to outperform Claude 3.7 at certain tasks and is the first non-reasoning model to use some mini reasonings within the main answer, which makes it particularly smart (for a non-CoT type of model). It is available on OpenRouter, provided by chutes.ai which seems to host their stuff on US-located servers.

Though it has a dedicated button outside, that leads to the Old UI's Sidebar.