The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability. It's also better at working with uploaded files, providing deeper insights & more thorough responses. GPT-4o ("o" for "omni") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of GPT-4 Turbo while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.
Evaluations
10
across 10 benchmarks
Latency
646ms
Context length
128k
tokens
Cost
$2.50 · $10
input · output per 1M tokens
Key takeaways
GPT-4o (2024-11-20) is a multimodal AI model by OpenAI, designed for text and image inputs with text outputs. It aims to enhance creative writing, file processing, and visual comprehension, delivering natural, engaging, and personalized responses. It functions as a general-purpose AI assistant with a focus on creative and analytical text generation.
This model is built on a proprietary Transformer-based architecture and features an extensive context length of 128,000 tokens. It is not open-weights.
GPT-4o (2024-11-20) excels in creative writing, processing uploaded files, and non-English language tasks. However, it struggles with complex mathematical reasoning, precise formatting, and advanced calculation accuracy, particularly in highly constrained or intricate problem-solving scenarios.
The model achieved 73.2% accuracy on the MATH-500 dataset, demonstrating strong logical reasoning in algebra, number theory, and prealgebra, especially in problems requiring step-by-step reasoning. It struggled with abstract counting and probability, complex multi-step problems, and strict mathematical formatting.
In the AIME 2025 dataset (advanced mathematics), GPT-4o scored 0% accuracy, consistently failing to produce correct final answers despite providing detailed step-by-step solutions. Its primary weaknesses were computational errors, misinterpretation of constraints, and incorrect application of advanced mathematical principles.
