OpenAI: gpt-oss-120b (free)

The gpt-oss-120b model from OpenAI is an open-weight, 117-billion-parameter Mixture-of-Experts (MoE) language model, specifically engineered for advanced reasoning, agentic applications, and general-purpose production environments. This powerful model activates 5.1 billion parameters per forward pass, making it highly efficient. It is optimized to run on a single H100 GPU, leveraging native MXFP4 quantization for superior performance. Key capabilities of gpt-oss-120b include configurable reasoning depth, providing users with granular control over its thought processes. It offers full chain-of-thought access, enhancing transparency and debuggability. Furthermore, the model boasts native tool use, encompassing essential features like function calling, web browsing, and structured output generation. With a generous 131K token context window and a maximum output of 4K tokens, it's ideal for complex interactions. Best of all, it's available for free on Multi AI, with $0.00 pricing per 1M input/output tokens. Explore its potential today!