Skip to content
IFAI

OpenAI Evals

Product2 stories

  1. AI toolsReach 52

    OpenAI launches Evals API to automate prompt testing

    OpenAI just rolled out its new Evals API, allowing developers to programmatically test prompts and model outputs instead of clicking through a web dashboard. Teams can write custom Python evaluation logic, plug in test datasets, and integrate checks directly into CI/CD pipelines. It basically turns LLM prompt evaluation into standard unit testing.

  2. AI modelsReach 57

    OpenAI drops GPT-4: Smarter, multimodal, and passes the bar

    OpenAI just launched GPT-4, a multimodal model that accepts text and image inputs while showing human-level benchmark performance. It scored in the top 10% of a simulated bar exam—trouncing GPT-3.5's bottom 10% showing—and cuts disallowed responses by 82%. Text capabilities are live for ChatGPT Plus and API waitlists, while image features remain in research preview.