OpenAI released ChatGPT Images 2.5 on 8 September 2026, advertising more precise local edits, better consistency across edits, more faithful reference products and sharper detail. We evaluate these claims on four forgery tasks with answers fixed in advance: receipt-field alteration, repeated editing, product placement...
Ankit Raj, Yu-Xin Zhang, Kidus Zewde et al.· 2 citations
AgentForge-Bench measures how reliably an off-the-shelf coding agent, driving one of seven open-weight models with a shell and the stock Python PDF stack, alters one dollar amount, date or address in a real filed financial document from a single sentence of intent, graded by rules rather than by a model.
Si-Miao Ren, Ankit Raj, Tommy Duong et al.· 2 citations
The frozen collection of ChatGPT Images 2.5 supports analysis of reported model use during a product transition, with source and attribution evidence retained for interpretation, and is released at https://scam.ai/research.
D. Ng, Xing-Yu Shen, Ankit Raj et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.