Show HN: Flint – A 30B model fine-tuned for less repetition
Posted by thmsmxwll 5 days ago
As frontier LLMs have very little output diversity even for open ended queries. We built Flint to see if we could reverse this. It’s a finetuned Qwen3 30B model specifically trained to produce higher entropy when asked open ended questions.
Flint significantly increases the NoveltyBench score compared to the base model, without significantly reducing the score on non-creative benchmarks like MMLU-STEM.
This shows that that divergence tuning doesn't actually have to be a tax on base capabilities.
Flint scores 7.47/10 on NoveltyBench while most frontier models score between 1.8 and 3.2.
Comments
Comment by Bolwin 4 days ago
Interesting. Would be able to release it on api or weights so we can use it outside the context of your application?
Comment by pipspringboards 4 days ago
@bolwin we will definately do that in future. Still in alpha and making updates and scaling it up but hopefully in next 3-6months we will have an API available