Ask HN: Is anyone doing anything cool with tiny language models?

Re:Ask HN: Is anyone doing anything cool with tiny language ... by pseudosavant on Tuesday January 21, 2025 @05:22PM (#42785838)

I wonder how big that model is in RAM/disk. I use LLMs for FFMPEG all the time, and I was thinking about training a model on just the FFMPEG CLI arguments. If it was small enough, it could be a package for FFMPEG. e.g. `ffmpeg llm "Convert this MP4 into the latest royalty-free codecs in an MKV."`

[ Parent | on HN ]

Re:Ask HN: Is anyone doing anything cool with tiny language ... by h0l0cube on Tuesday January 21, 2025 @06:10PM (#42786381)

Please submit a blog post to HN when you're done. I'd be curious to know the most minimal LLM setup needed get consistently sane output for FFMPEG parameters.

[ Parent | on HN ]

Re:Ask HN: Is anyone doing anything cool with tiny language ... by jedbrooke on Tuesday January 21, 2025 @05:33PM (#42785929)

the jetbrains models are about 70MB zipped on disk (one model per language)

[ Parent | on HN ]

Re:Ask HN: Is anyone doing anything cool with tiny language ... by pseudosavant on Wednesday January 22, 2025 @11:42AM (#42794671)

That is easily small enough to host as a static SPA web app. I was first thinking it would be cool to make a static web app that would run the model locally. You'd make a query and it'd give the FFMPEG commands.

[ Parent | on HN ]

Re:Ask HN: Is anyone doing anything cool with tiny language ... by binary132 on Tuesday January 21, 2025 @07:29PM (#42787136)

That’s a great idea, but I feel like it might be hard to get it to be correct enough

[ Parent | on HN ]

Re:Ask HN: Is anyone doing anything cool with tiny language ... by maujim on Tuesday January 21, 2025 @06:35PM (#42786629)

from a few days ago: https://news.ycombinator.com/item?id=42706637

[ Parent | on HN ]