Bug report

LLMs are greyed out and Merlin keeps saying I have exceeded usage limits even though I have over 2000 credits left. Also, on many screens the credits left is hidden?

The merlin extension for linked in has stopped working in the comments.

RAG tool is not giving results in Projects. It is completely broken. ChatGPT free account giving better results.

I keep uploading documents and asking for outputs based on my documents. But it gives outputs not relevant to my documents, and gives output from other people's documents. This is also a security concern - are my documents being sent to LLMs for output to other people, like theirs are to me? Happens with Gemini, Claude, GPT.

merlin chrome extension which was working fine till date now its not working on LinkedIn. I deleted and then reloaded Merlin extension and restarted the computer but still not working

I tried to chat with a TXT file, the LLM could not get the date using it. (Then it worked after switching to Google Gemini Flash Lite. Here's the "Thinking" part, retranslated to English (btw. I like the anti-hallucination efforts): ============================================ According to the system message, this is my last chance to answer this question. The device has already been called and the results are returned. Now I seem to get a generic error message from the RAG tool saying "Previously, rag was performed on the attachments for..." - this is someone's preprocessing response, not the content of the data. This means that the document was processed, but the data was not properly returned for analysis. The user sent me a request in Hungarian and wants me to analyze YouTube comments from a sociological perspective and quote interesting comments. Since I didn't get the actual content from the document, I have to be honest with the user about this. Based on the description of the document, I know that it is about the topic of Hungarian elections and political changes (Orbán vs. Magyar), but I cannot access the actual comments. According to the instructions: DO NOT generate new device calls Acknowledge your limitations If the device call failed, notify the user Be transparent about what I've achieved and what I haven't Therefore, it must be honestly admitted that during the processing of the document it was not possible to retrieve the current comments, and I cannot perform the requested sociological analysis.

Every day when I open the app on my android phone, it prompts me to log in. I use Bitwarden as my password manager, and it doesn't detect this app as being something to log into. Usually, a long press will give me an option to autofill that can then trigger Bitwarden to function on the page, but it doesn't recognize it as a login. I understand that is more a Bitwarden issue, but even after manually opening Bitwarden and copying the password, "paste" on the app is greyed out, both for the username and password fields. out of the 200+ apps I use on my phone, this is the only one that does this. Could you at least add biometrics to simplify the login process?

Using the new interface. The format of my prompt (Markdown etc.) always becomes a giant clumped text whenever I access them from saved prompt. Greatly impeding the prompting experience in the new interface

I am writing to express my disappointment regarding the fair usage limits applied to my account, which were not clearly communicated during the registration and purchase phase. Like several other users, for example, as mentioned in this feedback Merlin AI Pro Unlimited Plan Reality - 100 Monthly Cap, these limits do not feel like "fair" limits but rather sustainability limits that should have been clearly disclosed upfront. Nevertheless, I acknowledge this condition and will decide soon whether to continue using the service or to suspend my subscription and consider alternative providers. In the meantime, I would like to request a more detailed breakdown of the available credits on a daily and monthly basis, so I can better assess if the current plan fits my needs. It would be helpful to have clear, specific usage information such as: "You have used 700 credits out of 1000." Currently, the percentage progress bar shown in the "Account" section is not sufficiently clear. Thank you for your attention to this matter. I look forward to your prompt response. EDIT: I have exported the cURL request the application uses to retrieve the remaining quota for my user, so I am able to access this information independently. However, it is rather inconvenient and not accessible for those who are not programmers. Therefore, it would be "fair" to make this information publicly available to all users.

For past few days i am noticing there's quite difference in the quality of merlin and the original ai tools have chatgpt subscription also even with same Prompt the output is very different, and i am selecting the same model