Advanced models

Advanced models are available with Bring Your Own Key feature or Bring Your Own Model.

GPT-4o (BYOK)

GPT-4o ("o" for "omni") is OpenAI's flagship model designed for natural human-computer interaction, capable of processing and generating content across text, audio, image, and video with remarkable speed—responding to audio inputs in as little as 232 milliseconds. It matches GPT-4 Turbo's performance on English text and code while significantly improving multilingual capabilities, and excels particularly in vision and audio understanding compared to previous models. GPT-4o is ideal for real-time conversations, multilingual communication, multimodal content creation, audio transcription, and accessible AI applications where natural interaction is crucial. At half the cost of GPT-4 Turbo in the API with twice the speed, it represents a major advancement in making sophisticated AI capabilities more broadly accessible.

GPT-4.1 (BYOK)

GPT-4.1 is OpenAI's latest flagship model featuring exceptional performance in coding, instruction following, and long-context comprehension with a massive 1 million token context window. It outperforms GPT-4o across all benchmarks, with remarkable 21.4% improvement in coding tasks (54.6% on SWE-bench Verified) and 10.5% better instruction following abilities. GPT-4.1 excels at complex software engineering, analyzing large documents, processing entire codebases, and powering reliable AI agents for real-world tasks. With improved long-context understanding, more reliable formatting, and 26% lower cost than GPT-4o, it's ideal for developers building sophisticated applications requiring precision and reasoning across extensive inputs.