OpenAI has unveiled two cutting-edge open-weight language models, gpt-oss-120b and gpt-oss-20b, under the Apache 2.0 license. These models excel in advanced reasoning tasks and tool usage, and are optimized for efficient deployment on both consumer and professional hardware. The gpt-oss-120b model, with 117 billion parameters, can run on a single 80 GB GPU, while the gpt-oss-20b model, with 21 billion parameters, is suitable for personal computers with just 16 GB of memory. Both models deliver performance comparable to OpenAI's proprietary o3-mini and o4-mini models, and are particularly strong in coding, mathematical reasoning, and health-related tasks. This release marks OpenAI's first open-weight models since GPT-2 in 2019, offering developers greater control and flexibility to run and customize AI on their own infrastructure. ([openai.com](https://openai.com/index/introducing-gpt-oss?utm_source=openai))