HNHacker News
TopNewBestAskShowJobs

ibragim_bad

28 karma · joined May 29, 2025

submissionscomments

Opus 5.5 vs. GPT-6 in pi-agent: reasoning efforts and DeepSeek, GLM, Qwen

ibragim.dev·3 pts·ibragim_bad·
0

13 Models and 4 Agents on SWE Tasks: Go, Java, Python, Rust, TS

swe-rebench.com·51 pts·ibragim_bad·
15

SWE-Rebench (December 2025)

swe-rebench.com·1 pts·ibragim_bad·
0

SWE-rebench: Over 21,000 Open Tasks for SWE LLMs

huggingface.co·5 pts·ibragim_bad·
1