OpenAI rolls out GPT-4.1 models to ChatGPT

In a significant development, OpenAI has rolled out its GPT-4.1 models to ChatGPT, designed to enhance coding and instruction-following capabilities for software engineers.

The release of GPT-4.1 has drawn attention to AI coding tools, with OpenAI reportedly nearing a $3 billion acquisition of Windsurf, a popular AI coding tool.

The GPT-4.1 models outperform GPT-4o in coding tasks, scoring 54.6% on SWE-bench Verified, a measure of real-world software engineering skills, representing a 21.4% improvement over GPT-4o.

It should be noted that the GPT-4.1 models excel in several areas, including coding, instruction following, and long-context comprehension.

These models can process up to 1 million context tokens, making them suitable for tasks involving large codebases or lengthy documents.

The Sam Altman-led AI startup has claimed that GPT-4.1 is significantly better than GPT-4o at various coding tasks, such as agentically solving coding tasks, frontend coding, and producing code diffs in various formats.

The GPT-4.1 is now available to subscribers of ChatGPT Plus, Pro, and Team, while GPT-4.1 mini is available to free and paying users.

While OpenAI has faced criticism for releasing GPT-4.1 without a safety report, the company has argued that the model doesn't introduce new modalities or surpass o3 in intelligence, making its safety considerations different from frontier models.