OpenAI released GPT-OSS — two open-weight reasoning models you can run yourself: gpt-oss-20b and gpt-oss-120b. • 120b is close to o4-mini on reasoning, runs on a single 80GB GPU • 20b runs on a 16GB consumer GPU • You can run them via Transformers, vLLM, or Ollama • It’s not just a model drop — you get tooling, support, and agentic abilities • But: You’re responsible for safety, logging, and ops

Browser acts as ai agent

Briefly learn about the open source license, and apply the license to your source code uploaded on github 🙂