Harness bug masked Gemini 4 Argon: rerun with Antigravity beats Astra, nears Opus 5.5
vitaliychiley · x · 2026-10-27
Gemini 4 Argon appears underrated due to eval harness issues. While auditing traces, wenhaocha1 found Pi Harness simply ended the run whenever it hit a fallback (e.g. security tasks) in 20% of runs. Rerunning Argon with the Antigravity Harness pushed it past Astra to just behind Opus 5.5.
Separately, aiforsuccess reports Argon works extremely well with Antigravity, with Gemini models increasingly becoming their default across projects, and anticipates the rumored Carbon model.
More from coding & agent
- Claude spots broken physics in AI videos, fixes them with CFD and finite-element verification — OfirPress · 2026-10-27
- One prompt: Claude Opus 5.5 adds drone flight mode to RV travel game — ezshine · 2026-10-27
- Do canary tokens still work in the coding-agent era? Fake AWS keys may lure agents — kuza55 · 2026-10-27
- Veteran engineer on what the AI-coding generation will never feel: 36-hour sprints and elusive bugs — bendee983 · 2026-10-27
- Claude Opus 5.5 builds a browser open-world island by comparing screenshots to 341 real photos — prasenx · 2026-10-27
- BreakingWeb benchmark: browser agents fail about half of controlled task variants — shuyanzh36 · 2026-10-27