OpenCode 2.x notably better than 1.19, Hy4 trails Pi in tester's runs
mariofilhoml · x · 2026-11-12
RuneBench tester mariofilhoml shared observations from his coding-agent runs: OpenCode 2.x performs noticeably better than 1.19, and Hy4's OpenCode scored worse than Pi in his tests (run on Pi at xhigh). With no paid tier on OpenRouter, he'll finish the full run once his daily quota resets — a useful same-harness comparison for anyone picking coding agent tooling.
More from coding & agent
- Mote: open-source Markdown-to-DOCX/PPTX/PDF generator built for AI agents — kristoph · 2026-11-12
- xAI ships Grok Bot 1.0: AI teammates that join meetings and work in Slack — soleio · 2026-11-12
- Ten gems from Diogo's Code Summit talk: RLHF is part of the problem, software is back — TheTuringPost · 2026-11-12
- Google and Artificial Analysis launch MKA benchmark to verify agent deliverables — OnlineInference · 2026-11-12
- Hairline skill makes Perplexity Computer draw interactive isometric line figures from one line — jeffgrimes9 · 2026-11-12
- Cadpole: open-source parametric CAD where your coding agent builds, measures and self-checks models — MikeLewis · 2026-11-12