\n<\/p>\n
On Thursday, OpenAI released GPT-5.4<\/a>, a new foundation model billed as \u201cour most capable and efficient frontier model for professional work.\u201d In addition to the standard version, GPT-5.4 is also available as a reasoning model (GPT-5.4 Thinking) or optimized for high performance (GPT-5.4 Pro).<\/p>\n The API version of the model will be available with context windows as large as 1 million tokens, by far the largest context window available from OpenAI.<\/p>\n OpenAI also emphasized improved token efficiency, saying GPT-5.4 was able to solve the same problems with significantly fewer tokens than its predecessor.<\/p>\n The new model comes with significantly improved benchmark results, including record scores in computer use benchmarks OSWorld-Verified and WebArena Verified. The new model also scored a record 83% on OpenAI\u2019s GDPval test for knowledge work tasks.<\/p>\n GPT-5.4 also took the lead on Mercor\u2019s APEX-Agents benchmark<\/a>, designed to test professional skills in law and finance, according to a statement from Mercor CEO Brendan Foody.<\/p>\n \u201c[GPT-5.4] excels at creating long-horizon deliverables such as slide decks, financial models, and legal analysis,\u201d Foody said in the statement, \u201cdelivering top performance while running faster and at a lower cost than competitive frontier models.\u201d<\/p>\n GPT-5.4 continues the company\u2019s efforts to limit hallucinations and factual errors. OpenAI said the new model was 33% less likely to make errors in individual claims when compared to GPT 5.2, and overall responses were 18% less likely to contain errors.<\/p>\n Techcrunch event<\/p>\n \n\t\t\t\t\t\t\t\t\tSan Francisco, CA<\/span> As part of the launch, OpenAI has reworked how the API version of GPT-5.4 manages tool calling, introducing a new system called Tool Search. Previously, system prompts would lay out definitions for all available tools when calling the model \u2014 a process that could consume a lot of tokens as the number of available tools grew. The new system allows models to look up tool definitions as needed, resulting in faster and cheaper requests in systems with many available tools.<\/p>\n OpenAI has also included a new safety evaluation<\/a> to test its models\u2019 chain-of-thought, the running commentary given by the models to show thought process through multi-step tasks. AI safety researchers have long worried that reasoning models could misrepresent their chain-of-thought, and testing shows<\/a> it can happen under the right circumstances.<\/p>\n OpenAI\u2019s new evaluation shows that deception is less likely to happen in the Thinking version of GPT-5.4, \u201csuggesting that the model lacks the ability to hide its reasoning and that CoT monitoring remains an effective safety tool.\u201d<\/p>\n<\/div>\n
\n\t\t\t\t\t\t\t\t\t\t\t\t\t|<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\tOctober 13-15, 2026<\/span>\n\t\t\t\t\t\t\t<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n