News for hackers. Stuff that matters.

A comment on Open models by OpenAI. Parent comment

Re:Open models by OpenAI by itake on Wednesday August 06, 2025 @02:11AM (#44808299)
Local micro models are both fast and cheap. We tuned small models on our data set and if the small model thinks content is a certain way, we escalate to the LLM.

This gives us really good recall at really low cloud cost and latency.

[ Parent | on HN ]
Re:Open models by OpenAI by bavell on Wednesday August 06, 2025 @08:30AM (#44811096)
I'd love to try this on my data set - what approach/tools/models did you use for fine-tuning?
[ Parent | on HN ]
Re:Open models by OpenAI by itake on Wednesday August 06, 2025 @10:01PM (#44819836)
Everything is built in-house unfortunately. Many of our small models are turned Qwen3. But we mostly chose the model on SOTA at the time we needed a model trained.
[ Parent | on HN ]