Antigravity

🔑 Semi-Open

AI-native development platform with autonomous agent workflows, multi-step task planning, and integrated deployment pipelines.

💰 Beta access — pricing TBD · web, api

70.5Overall Score

Non-Gameable Scoring

Scores are derived from established benchmarks, adjusted for harness-specific performance across four dimensions: Coding, Reasoning, Tool Use, and Autonomy.

Each dimension starts from public benchmark data and applies harness-specific modifiers based on tool integration, context handling, and orchestration quality. The overall score is a weighted composite that penalizes narrow optimization.

Model	Overall	Coding	Reasoning	Tool Use	Autonomy
Claude Opus 4.6	70.5	70.68	76.14	61.35	88.75
Gemini 3.1 Pro	60.4	75.44	73.07	52.86	75.03
GPT-5.4	55.5	45.32	56.28	75.93	79.56
Claude Sonnet 4.6	54.8	62.7	47.09	57.19	87.34
Gemini 3 Pro	42.2	41.18	47.95	41.55	58.78
GPT-5.2	38.0	48.29	54.79	36.67	50.12
Gemini 3 Flash	36.7	48.69	46.9	29.84	41.21
GPT-5 (high)	35.6	50.69	44.12	28.96	54.22
Claude Sonnet 4.5	35.6	47.37	38.94	28.88	62.62
Claude Opus 4.5	34.5	50.7	38.99	21.54	61.14