GPT-4.1

GPT-4.1 is OpenAI's API-focused model released in April 2025, representing a major leap in context length capabilities with a 1 million token context window-nearly 8x the capacity of GPT-4o's 128K tokens. This massive context window enables the model to handle extremely long documents, codebases, and multi-turn conversations with minimal information loss.

The model achieves 38.3% on the MultiChallenge benchmark for instruction following, surpassing GPT-4o by 10.5 percentage points. It has been specifically optimized for long-context tasks, coding workflows, and complex instruction following, making it particularly valuable for enterprise applications that require processing extensive documentation or maintaining context across lengthy interactions.

GPT-4.1 is available in multiple variants including GPT-4.1 mini and GPT-4.1 nano, offering different trade-offs between capability, speed, and cost. The model's extended context window and improved instruction-following capabilities make it a strong choice for applications requiring deep contextual understanding over long sequences.