Introducing GPT-5.2

We are introducing GPT‑5.2, the most capable model series yet for professional knowledge work.

Already, the average ChatGPT Enterprise user says⁠ AI saves them 40–60 minutes a day, and heavy users say it saves them more than 10 hours a week. We designed GPT‑5.2 to unlock even more economic value for people; it’s better at creating spreadsheets, building presentations, writing code, perceiving images, understanding long contexts, using tools, and handling complex, multi-step projects.

GPT‑5.2 sets a new state of the art across many benchmarks, including GDPval, where it outperforms industry professionals at well-specified knowledge work tasks spanning 44 occupations.

Notion⁠(opens in a new window), Box⁠(opens in a new window), Shopify⁠(opens in a new window), Harvey⁠(opens in a new window) and Zoom⁠(opens in a new window) observed GPT‑5.2 demonstrates state-of-the-art long-horizon reasoning and tool-calling performance. Databricks⁠(opens in a new window), Hex⁠(opens in a new window) and Triple Whale⁠(opens in a new window) found GPT‑5.2 to be exceptional at agentic data science and document analysis tasks. Cognition⁠(opens in a new window), Warp⁠(opens in a new window), Charlie Labs⁠(opens in a new window), JetBrains⁠(opens in a new window) and Augment Code⁠(opens in a new window) say GPT‑5.2 delivers state-of-the-art agentic coding performance, with measurable improvements in areas such as interactive coding, code reviews and bug finding.

In ChatGPT, GPT‑5.2 Instant, Thinking, and Pro will begin rolling out today, starting with paid plans. In the API, they are available now to all developers.

Overall, GPT‑5.2 brings significant improvements in general intelligence, long-context understanding, agentic tool-calling, and vision—making it better at executing complex, real-world tasks end-to-end than any previous model.

GPT‑5.2 Thinking is the best model yet for real-world, professional use. On GDPval⁠, an eval measuring well-specified knowledge work tasks across 44 occupations, GPT‑5.2 Thinking sets a new state-of-the-art score, and is our first model that performs at or above a human expert level. Specifically, GPT‑5.2 Thinking beats or ties top industry professionals on 70.9% of comparisons on GDPval knowledge work tasks, according to expert human judges. These tasks include making presentations, spreadsheets, and other artifacts. GPT‑5.2 Thinking produced outputs for GDPval tasks at >11x the speed and <1% the cost of expert professionals, suggesting that when paired with human oversight, GPT‑5.2 can help with professional work. Speed and cost estimates are based on historical metrics; speed in ChatGPT may vary.

When reviewing one especially good output, one GDPval judge commented, "It is an exciting and noticeable leap in output quality... [it] appears to have been done by a professional company with staff, and has a surprisingly well designed layout and advice for both deliverables, though with one we still have some minor errors to correct."

Additionally, on our internal benchmark of junior investment banking analyst spreadsheet modeling tasks—such as putting together a three-statement model for a Fortune 500 company with proper formatting and citations, or building a leveraged buyout model for a take-private—GPT 5.2 Thinking's average score per task is 9.3% higher than GPT‑5.1’s, rising from 59.1% to 68.4%.

Side-by-side comparisons show improved sophistication and formatting in spreadsheets and slides generated by GPT‑5.2 Thinking:

To use the new spreadsheet and presentation capabilities in ChatGPT, you must be on a Plus, Pro, Business, or Enterprise plan and select either GPT‑5.2 Thinking or Pro. Complex generations can take many minutes to produce.

GPT‑5.2 Thinking sets a new state of the art of 55.6% on SWE-Bench Pro, a rigorous evaluation of real-world software engineering. Unlike SWE-bench Verified, which only tests Python, SWE-Bench Pro tests four languages and aims to be more contamination-resistant, challenging, diverse, and industrially relevant.

On SWE-bench Verified (not plotted), GPT‑5.2 Thinking scores our new high of 80%.

For everyday professional use, this translates into a model that can more reliably debug production code, implement feature requests, refactor large codebases, and ship fixes end-to-end with less manual intervention.

GPT‑5.2 Thinking is also better at front-end software engineering than GPT‑5.1 Thinking. Early testers found it significantly stronger at front-end development and complex or unconventional UI work—especially involving 3D elements—making it a powerful daily partner for engineers across the stack. See a few examples of what it can produce from a single prompt:

Early testers shared their feedback on GPT‑5.2’s coding capabilities:

GPT‑5.2 Thinking hallucinates less than GPT‑5.1 Thinking. On a set of de-identified queries from ChatGPT, responses with errors were 30%_rel less common. For professionals, this means fewer mistakes when using the model for research, writing, analysis, and decision support—making the model more dependable for everyday knowledge work.

Like all models, GPT‑5.2 Thinking is imperfect. For anything critical, double check its answers.

GPT‑5.2 Thinking sets a new state of the art in long-context reasoning, achieving leading performance on OpenAI MRCRv2—an evaluation that tests a model’s ability to integrate information spread across long documents. On real-world tasks like deep document analysis, which require related information across hundreds of thousands of tokens, GPT‑5.2 Thinking is substantially more accurate than GPT‑5.1 Thinking. In particular, it’s the first model we’ve seen that achieves near 100% accuracy on the 4-needle MRCR variant (out to 256k tokens).

In practical terms, this enables professionals to use GPT‑5.2 to work with long documents—such as reports, contracts, research papers, transcripts, and multi-file projects—while maintaining coherence and accuracy across hundreds of thousands of tokens. This makes GPT‑5.2 especially well suited for deep analysis, synthesis, and complex multi-source workflows.

For tasks that benefit from thinking beyond the maximum context window, GPT‑5.2 Thinking is compatible with our new Responses /compact endpoint, which extends the model’s effective context window. This lets GPT‑5.2 Thinking tackle more tool-heavy, long-running workflows that would otherwise be limited by context length. Read more in our API documentation⁠(opens in a new window).

GPT‑5.2 Thinking is our strongest vision model yet, cutting error rates roughly in half on chart reasoning and software interface understanding.

For everyday professional use, this means the model can more accurately interpret dashboards, product screenshots, technical diagrams, and visual reports—supporting workflows in finance, operations, engineering, design, and customer support where visual information is central.

Compared to previous models, GPT‑5.2 Thinking has a stronger grasp of how elements are positioned within an image, which helps on tasks where relative layout plays a key role in solving the problem. In the example below, we ask the model to identify the components in an image input (in this case, a motherboard) and return labels with approximate bounding boxes. Even on a low-quality image, GPT‑5.2 identifies the main regions and places boxes that sometimes match the true locations of each component, while GPT‑5.1 only labels a few parts and shows a much weaker understanding of their spatial arrangement. Both models make clear mistakes, but GPT‑5.2 shows better comprehension of the image.