←back to thread

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

684 points prettyblocks | 5 comments | 21 Jan 25 19:39 UTC | HN request time: 0.647s | source

I mean anything in the 0.5B-3B range that's available on Ollama (for example). Have you built any cool tooling that uses these models as part of your work flow?
1. sidravi1 ◴[22 Jan 25 03:04 UTC] No.42788468[source]▶
We fine-tuned a Gemma 2B to identify urgent messages sent by new and expecting mothers on a government-run maternal health helpline.

https://idinsight.github.io/tech-blog/blog/enhancing_materna...

replies(4): >>42788954 #>>42790308 #>>42793587 #>>42801392 #
2. proxygeek ◴[22 Jan 25 04:19 UTC] No.42788954[source]▶
Such a fun thread but this is the kind of applications that perk up my attention!

Very cool!

3. Mashimo ◴[22 Jan 25 08:09 UTC] No.42790308[source]▶
Oh that is a nice writeup. We have something similar in mind at work. Will forward it.
4. Mumps ◴[22 Jan 25 15:06 UTC] No.42793587[source]▶
lovely application!

Genuine question: why not use (Modern)BERT instead for classification? (Is the json-output explanation so critical?)

5. Mukina ◴[23 Jan 25 06:53 UTC] No.42801392[source]▶
Super cool. What a simple and powerful way to help mothers in need. Thanks for sharing.