Have something to say?
Append AI-generated text to original filename
Please consider adding an option to append the AI-generated output to the original filename. Use case: I use Written Instructions to analyze the content of the image, choose keywords that describe it, and append keywords at the end of the original unchanged filename. It seems not to work no matter what model I choose or what prompt I write, I cannot get the original filename. It seems like in Written Instructions the model has no access to the original filename and ignores all related instructions. Suggested solutions: Pass the original filename to the model Add Filename format section after Instructions in Written Instructions Thanks for considering, J
MLX AI Model selection Not Working
when I Download one of the available MLX models to use on my Mac. It downloads & when I hit Start MLX, it does nothing. When I try to run an operation, it wants an API key for OpenAI. Is there no setting to select which model to use anymore? If I have three different API keys for Anthropic, Gemini, & OpenAI, how do I select the one I want to use at that moment in MLX? It's defaulting automatically & the MLX feature isn't working. Is there a fix? Somebody figured it out? I'm currently updated
Feature request: ability to filter files to process
Example use cases: when user selects a folder with files (s)he may want to rename only some files and leave other types of files unchanged: - rename only PDFs - rename files beginning with some sequence . 🧑🔬
Add clear explanations for interface features
It might be beneficial to provide explanations for all available features. Some features seem to be explained neither in the interface nor in the online docs. For instance (see attached image), what should one write in the field that is “Used only when the field is required”? Should it be left blank? Then what is it for? 🧑🔬
Feature suggestion: Allow extracting more fields than are used for renaming
It might improve accuracy if we could instruct the LLM to extract several fields and then use only a subset for renaming. For instance, extract: First_Author_Last_Name, First_Author_First_Name, Year, Title, Subtitle, First_sentence_of_the_first_paragraph, Header, and then use for naming the file only: [YYYY]_[First_Author_Last_Name]_[Title]. This might improve accuracy. In my sample of PDFs LLMs struggle with extracting Title when a subtitle or a name of the series is close to the title (incorrectly inferred title 20-30% of the time). If I am not mistaken, currently the “unused” fields to extract are being automatically removed from a template. 🧑🔬
Feature suggestion: preview of LLM "thinking" for debugging Presets
It would be super helpful to get a sneak peek of what an LLM decides step by step or what it is “thinking”. Right now, we’re debugging Presets “in the dark”. It seems like some templates or custom prompts aren’t working on certain files no matter what (for now, I tested on PDFs only). 🧑🔬
Feature suggestion: Refresh for "To Process"/"Processed" Contents
Add a Refresh function for “To Process”, either automatic or manual (eg., Command-R). Use case: NameQuick renamed some files. User manually adjusts new names in the Finder. However, all these files previously renamed by NameQuick are already in “Processed” folder and their names displayd in NameQuick do not change to manually renamed ones.
Bug report: infinite loop and memory explosion
Hi, not sure if this is the place to report bugs (if not, direct me to the right one please). NameQuick stopped responding and cased other apps to Pause because it used over 600GB. I traced it to a memory leak (Physical footprint: 622.3G). See lines 24-45 in attached process sample and lines 383-422 in diagnostic log. Hope this is helpful.
[BUG] Add better connection testing and troubleshooting for BYOK API configuration
All four APIs say they’re properly connected, but only Gemini actually works. No idea how to start troubleshooting here. It’d be helpful to have more robust connection testing and suggestions when it fails.
Auto-recognize sets of similar files for more helpful naming schema
There’s a bunch of scenarios where naming is required to disambiguate within a set of similar files — think a folder of invoices. It’d be great if QuickName could either have a “this is a set” type button when bulk selecting, or, better, do some quick checks for potential sets (e.g. items within the same folder, or files that are all the same type and nearly the same size). With sets, a great way to name is by front-loading the differentiating factor, say by listing the month + year first for a folder of bills. Effectively, help people see the sequence by scanning the first couple words.
Extract Data into Table or Sheet
Convert unstructured file content into structured rows and columns. The feature reads the selected files, identifies key fields, and outputs a clean table that users can export or continue editing. Example Use Cases: • Turn invoices into a bookkeeping sheet. • Collect line items from receipts. • Extract headings, authors, and keywords from academic papers.
Custom Prompts not visible
Using NameQuick with BYOK - I cannot see the mentioned Option “Settings → AI → Manage Custom Prompts”
Max. File Name Length
Currently, file names are cut off hard after approx 100 letters. How about incorporating the max file name length in the original prompt to avoid artificially cut off file names…?
Ollama processing error
I have Ollama configured, using model gemma3:12b - unfortunately when starting a smart renaming, the following error occurs: Processing failed: AI analysis failed: The feature 'enableLocalLLMPDFProcessing' is currently disabled. Is there something I am missing? Thanks for your help! Ollama and NameQuick - latest version installed MacOS - latest Tahoe public beta
怎么写的增加deepseek，正在做，又取消了？
目前谷歌，chatgpt，claude的api，在中国都不能用啊，增加个openrouter或者自定义代理比如closeai的支持也行啊，买了几个月了，一直瘫痪中