Hacker News·1 min read·hard
DeepSeek V4 Flash 0731
T
tosh
✦AI Summary
DeepSeek has released performance metrics for its V4 Flash 0731 model on the ARC-AGI benchmarks. The model achieved specific success rates at low cost-per-task intervals.
At max effort, DeepSeek V4 Flash 0731 scores 89.0% on ARC-AGI-1 Semi-Private at $0.02 per task and 61.4% on ARC-AGI-2 Semi-Private at $0.04 per task.
technologyai
✦
Get the full story
Sign up for Headlinne to unlock AI insights, political bias analysis, and your personalized news feed.
Create free accountAlready have an account? Sign in