GLM-4.7-Flash Hits 59.2% on SWE-Bench With Only 3B Active Parameters
A new open-source coding model from DeepReinforced achieves competitive benchmark scores at a fraction of the size of leading models — small enough to run on a laptop.
DeepReinforced's InterX framework has produced GLM-4.7-Flash, an open-source coding model that reportedly scores 59.2% on SWE-Bench while using only 3 billion active parameters, as reported by @testingcatalog. The model was also hyped by @itsPaulAi, who emphasized it runs on laptop hardware and is fully open source.
Unlock the full briefing
Get every story in today's briefing, the full archive, and the daily AI intelligence brief.
All stories today
Full archive
Daily brief
Cancel anytime. Payments powered by Stripe.