GPT-4.1

GPT-4.1 is a family of large language models launched by OpenAI in April 2025, available exclusively through the API. The series includes three models: GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano, offering significant improvements over GPT-4o and GPT-4o mini in coding, instruction following, and long-context comprehension. All models support up to 1 million tokens of context and feature a knowledge cutoff of June 2024. GPT-4.1 achieves state-of-the-art results on benchmarks such as SWE-bench Verified (54.6%), MultiChallenge (38.3%), and Video-MME (72.0% on long, no subtitles). The mini model balances performance with 83% cost reduction and nearly half the latency of GPT-4o, while the nano model is the fastest and cheapest, excelling at classification and autocompletion. The family is optimized for real-world developer tasks, including building reliable AI agents, software engineering, document analysis, and customer request resolution. GPT-4.1 is not available in ChatGPT; its improvements will be gradually incorporated into the ChatGPT version of GPT-4o.