Details
- One week after GPT-5.5 launch on April 23, 2026, OpenAI calls it their strongest model release yet.
- API revenue grows over 2x faster than any prior release, fueled by enterprise demand.
- Codex, OpenAI's coding tool, doubled revenue in under seven days due to rising need for agentic coding.
- GPT-5.5 excels in benchmarks like Terminal-Bench 2.0 (82.7%) and FrontierMath, outperforming rivals like Claude Opus 4.7 and Gemini 3.1 Pro per OpenAI data.
- Model targets complex tasks in coding, research, data analysis, scientific workflows, and drug discovery; available to paid users with API access starting April 24.
- Independent tests mixed: ZDNET praises improvements in writing, coding, reasoning over predecessors; Tom's Guide found it trailing Claude Opus 4.7 in 7 categories.
Impact
GPT-5.5's explosive revenue surge, with API growth over 2x prior releases and Codex doubling in a week, underscores surging enterprise adoption of agentic AI tools, pressuring rivals like Anthropic's Claude and Google's Gemini to match in coding and research benchmarks where OpenAI claims leads. This accelerates market shift toward paid, high-capability models for knowledge work and science, potentially widening OpenAI's lead in monetizing frontier AI amid intensifying competition.
