nebius
/

SWE-rebench-openhands-Qwen3-30B-A3B

Text Generation

Model card Files Files and versions

ibragim-bad commited on 4 days ago

Commit

2b25b52

·

verified ·

1 Parent(s): faefe64

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -164,7 +164,7 @@ When evaluated with the OpenHands (v0.54.0) agent, our 30B model:
 * Matches or surpasses the specialized Qwen3-Coder-30B-A3B-Instruct baseline at 500 turns on SWE-bench Verified (**50.3%** vs **50.0% Pass@1**; **68.3%** vs **63.0% Pass@5**).
 * Generalizes to longer interaction horizons, despite training on trajectories capped at 100 turns.
-For more details see our report in [Nebius blog](LINK-TO-BE-ADDED).
 ---

 * Matches or surpasses the specialized Qwen3-Coder-30B-A3B-Instruct baseline at 500 turns on SWE-bench Verified (**50.3%** vs **50.0% Pass@1**; **68.3%** vs **63.0% Pass@5**).
 * Generalizes to longer interaction horizons, despite training on trajectories capped at 100 turns.
+For more details see our report in [Nebius blog](https://nebius.com/blog/posts/openhands-trajectories-with-qwen3-coder-480b).
 ---