HN Explorer

sijuntan

↗ HN profile [ 23.3 ms ] full profile
Karma
0
Created
()
Submissions
9,669
  1. 1 points 0
  2. DeepSWE: Training an Open-Sourced Coding Agent by Scaling RL (pretty-radio-b75.notion.site)
    3 points 1
  3. DeepCoder: An Open-Sourced 14B Coder at O3-Mini Level (github.com)
    8 points 1
  4. DeepCoder: An Fully Open-Source 14B Coder at O3-Mini Level (pretty-radio-b75.notion.site)
    15 points 6
  5. DeepScaleR: Surpassing O1-Preview with a 1.5B Model by Scaling RL (pretty-radio-b75.notion.site)
    322 points 136