Feature Requests

Now that OpenAI has officially lowered the pricing for o3 — • $2.00 / 1M tokens (input) • $0.50 / 1M cached input • $8.00 / 1M tokens (output) — it makes no sense to keep o3 capped at 5 uses per day. If it’s now that affordable and it’s their most powerful reasoning model, then Merlin should absolutely unlock unlimited (or significantly higher) usage ASAP. The cost barrier is no longer a valid excuse. Thank you!

when i first came to merlin there were myriad models to choose from now you only have corporate giants. please bring in more open source models and lesser known models!

Hi! OpenAI has just released o3-pro, described as their most powerful model to date ( https://platform.openai.com/docs/models/o3-pro ). It offers leading performance across reasoning, math, code, and vision — making it ideal for both everyday tasks and complex workflows. Would it be possible to add o3-pro to Merlin? Having access to it would be a huge upgrade for power users looking for top-tier performance.

having uploads in a chat automatically saved in the vault makes it crazy messy and full of redundant files that i cant even sort

I will often request new models or checkpoints that are just released to be added to Merlin, especially if they are more capable and cheaper. However, it's unclear what is the exact models we have access to if there are any updates or checkpoints. I understand that a clean UI is important, but whether the specific model name and release data is shown in the model selection or elsewhere on the site, it's very helpful for your customers to know what is the exact model they are using. Can you please introduce more clarity on the exact models we have access to in Merlin? Thank you very much.

I want to update voice feature and create video, images then update the feature to upload photos on the device.

Add the o4-mini-high model. It’s already available for quite a long time.

Please update the descriptions of the models. (see attached)

Folders under History is great, but it is a pain when the folder collapses, which it always does. So suddenly you have a ton of chats under that folder you've got to scroll through. This also makes it laborious when trying to move chats out of the normal chat history into a folder, because the folders collapse as you scroll past them. It would be more efficient if you could select the chat and then select the folder instead of picking up the chat and manually relocating it.

The api can also modify images using prompts, can you add that to merlin itself as an option? Gpt-Image-01 works great in bonkers, but it is full capabilities isn't use well if it's only usable as a image generator. It's the most important part of gpt's-image-1, able to instruct and modify via text or instructed images Like these for example, as directly advertised by openai: https://youtu.be/GAf0KHM4fnk With that, it can then be also used for mobile users, prompting is the greatest advantage to gpt image generation Removing that function makes it look like a generic image generator that are only a bit better to other image generator