OpenAI Unveils GPT-5.2 to Compete with Google and Anthropic

OpenAI Unveils GPT-5.2 to Compete with Google and Anthropic

Key Points

  • OpenAI released GPT-5.2 with Instant, Thinking, and Pro versions.
  • GPT-5.2 outperformed GPT-5.1 on ten benchmarks, including a perfect AIME 2025 score.
  • Error rate reduced by 30 percent, improving factual accuracy.
  • Designed to excel at spreadsheets, presentations, code, image perception, and multi‑step tasks.
  • Positioned as a direct competitor to Google Gemini 3 Pro and Anthropic models.
  • Available exclusively on paid plans; GPT-5.1 remains for a limited period.
  • Internal “code red” effort spurred the rapid development of GPT-5.2.

OpenAI releases GPT-5.2 to take on Google and Anthropic

OpenAI Announces GPT-5.2

OpenAI introduced GPT-5.2, releasing three versions—Instant, Thinking, and Pro—designed for professional workloads. The company highlighted the model’s enhanced abilities in creating spreadsheets, building presentations, writing code, perceiving images, understanding long contexts, using tools, and managing complex, multi‑step projects.

Benchmark Performance

In a series of ten benchmarks, GPT-5.2 Thinking surpassed GPT-5.1, achieving a perfect 100 percent score on the AIME 2025 test, compared with GPT-5.1’s 94 percent. On the ARC‑AGI‑1 benchmark, the new system beat its predecessor by more than ten percentage points. OpenAI also reported a 30 percent reduction in factual errors, describing the model as more dependable for research, writing, analysis, and decision support.

Competitive Positioning

OpenAI frames GPT-5.2 as a direct response to Google’s Gemini 3 Pro and Anthropic’s models, aiming to reclaim top rankings on evaluation platforms such as LMArena, where Gemini 3 Pro recently led. The release follows internal “code red” efforts to improve performance after previous releases fell short of expectations.

Availability and Pricing

All three GPT-5.2 variants are initially available only to users on OpenAI’s paid plans. GPT-5.1 will remain accessible for a limited time, allowing paid users to select the legacy model for the next three months.

Implications for the AI Landscape

The launch underscores OpenAI’s commitment to advancing large‑language‑model capabilities and competing aggressively in a market increasingly dominated by Google and Anthropic. By emphasizing higher accuracy, broader tool use, and stronger multi‑step reasoning, OpenAI hopes GPT-5.2 will become the preferred choice for professional and enterprise users.

Source: engadget.com