OpenAI has announced the release of its new GPT-5.5 model, which the company describes as its most advanced and user-friendly model to date. This development marks a significant evolution in the capability of AI tools for complex tasks.
GPT-5.5 follows the release of GPT-5.4 last month and is tailored for improved efficiency in diverse functions such as coding, online research, and document preparation. The model is designed to handle multi-part tasks autonomously, enabling users to trust it with more intricate workflows.
OpenAI stated, “Instead of carefully managing every step, you can give GPT-5.5 a messy, multi-part task and trust it to plan, use tools, check its work, navigate through ambiguity, and keep going.” This flexibility is complemented by what the company claims is the strongest set of safety features to date and a capacity for reduced token usage in Codex applications.
The rollout of GPT-5.5 begins Thursday and will be available to Plus, Pro, Business, and Enterprise ChatGPT tiers, along with Codex users. A specialized GPT-5.5 Pro variant will be accessible to Pro, Business, and Enterprise customers specifically.
| Benchmark | GPT-5.5 | GPT-5.4 | GPT-5.5 Pro | GPT-5.4 Pro | Claude Opus 4.7 | Gemini 3.1 Pro |
|---|---|---|---|---|---|---|
| Terminal-Bench 2.0 | 82.7% | 75.1% | – | – | 69.4% | 68.5% |
| Expert-SWE (Internal) | 73.1% | 68.5% | – | – | – | – |
| GDPval (wins or ties) | 84.9% | 83.0% | 82.3% | 82.0% | 80.3% | 67.3% |
| OSWorld-Verified | 78.7% | 75.0% | – | – | 78.0% | – |
| Toolathlon | 55.6% | 54.6% | – | – | – | 48.8% |
| BrowseComp | 84.4% | 82.7% | 90.1% | 89.3% | 79.3% | 85.9% |
| FrontierMath Tier 1–3 | 51.7% | 47.6% | 52.4% | 50.0% | 43.8% | 36.9% |
| FrontierMath Tier 4 | 35.4% | 27.1% | 39.6% | 38.0% | 22.9% | 16.7% |
| CyberGym | 81.8% | 79.0% | – | – | 73.1% | – |
The introduction of GPT-5.5 further intensifies the competitive landscape between OpenAI and Anthropic, which also aims to go public later this year. Anthropic recently launched Claude Opus 4.7 and introduced Mythos Preview, a cybersecurity model. In response, OpenAI has released GPT-5.4-Cyber, focusing on identifying security vulnerabilities.
Both companies are vying to capture the market in AI coding and enterprise tools, with OpenAI prioritizing larger revenue-generating opportunities over less critical projects. This latest product announcement comes just days before a major trial involving Elon Musk and OpenAI executives Sam Altman and Greg Brockman, set to begin on Monday in Oakland, California.
