GLM-5.3 Flash trails full model by 5.6 points at 17x lower cost
Together AI ran 900 DeepSWE rollouts on GLM-5.3 and GLM-5.3 Flash. Flash trails by 5.6 points at pass@1 but only 2.6 at pass@4, at $0.24 vs $3.99 per rollout. A cascade starting with Flash solves 80.9% of tasks at $1.70 each.
How this story unfolded
4 days · 2 reports · 2 community posts · from Aug 29
- Aug 29
- Aug 30
- Sep 1
AI Models by email
Get an email when there's news on AI Models
No news that day, no email.
More stories today
- Claude Code 2.1.258 fixes macOS 12 launch regression
- AfterQuery reportedly becomes Y Combinator's fastest-ever unicorn at $3.2B
- Ethan Mollick: agents now capable of long-running self-organized work
- FDA builds AI-ready data foundation on Databricks for Government
- AWS and Amazon MGM Studios launch gen AI film fund