Synapse Open Synapse
Open source

Show HN: Kvark Sparkle 27B – reasoning model trained to think with fewer tokens

Show HN · · 4 min read

We had problems with Qwen spending too much time reasoning about tasks instead of doing them. That's why we decided to fine-tune it to use fewer reasoning tokens without losing accuracy. After a bunch of experiments we ended up with KVARK Sparkle 27B.

Latest from Show HN