I know this isn’t a question, but more of a general observation about LMs. However, I’d still like to say that a fine-tuned Qwen3 0.6B model can produce more effective and faster results than a raw Gemma 3 12B model. Maybe it’s because I’m not a programmer, but I believe being able to give commands in natural language adds a great deal of flexibility to software.
blenderq --blend /path/to/file1.blend /path/to/file2.blend love this option
normaly I just use blender -b path/file1 -a && blender -b path/file2 -a
thank you for making this I did not read your code and I'm curious about where did you get (or calculate) the progress value, thank you for making this
Thank you so much, actually I tried to run mistral and TinyLlama over ollama without success.
But I've never limit inferencing process or anything. Let me try LocalAI
most of the people I know use WhatsApp and Telegram because they can read their notes from any device/ant time and can forward it any one.
so for this use case ideabook cannot do what WhatsApp or Telegram do for me.
But I like the clean interface.
note: mobile appstore links are not there
https://github.com/umtksa/links (repo)
https://umtksa.github.io/links/ (demo)