AI Coding Daily (Antigravity) - Total: leaderboard

AI Coding Daily through the Antigravity harness: the publisher’s displayed total across code-quality, bug-finding and edge-case tests, up to 90 points, after deductions for inconsistent results across attempts. Higher is better. This score is read directly, not reconstructed by summing component boards; model and reasoning-effort labels remain separate.

Metric: Total points (max 90). Source: aicodingdaily.com. Saturation forecast: Around February 2027. 2 models tracked. Updated 20 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 4 NeonHigh79.85#3
2Gemini 4 HydrogenHigh78.01#16

No result here: #1 Claude Fable 5.2, #4 GPT-6.2 Astra, #5 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=ai-coding-daily-antigravity-total · How It Works · Data refreshed daily, snapshot 2026-11-20.