gpt-oss-120b (free)
🧠 AI Modelopenai
117B MoE model with 5.1B active params per forward, free and open-weight on OpenRouter.
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model by OpenAI. It activates only 5.1B parameters per forward pass, achieving high efficiency while maintaining strong reasoning capabilities. The model supports a context length of 131,072 tokens and handles text-only inputs and outputs. It includes features such as reasoning, seed control, temperature, stop sequences, and tool calls. Benchmarks show ELO ratings of 981 for 3D tasks and 1015 for code categories. Designed for high-reasoning, agentic, and general-purpose production use cases, it is available for free via OpenRouter, making advanced AI accessible to developers and researchers.
💡Highlights
- ├─117B MoE, only 5.1B active per token
- ├─131k context length available free
- └─OpenAI open-weight, reasoning-focused model
🎯For
- ├─AI researchers
- ├─Developers
- └─Production engineers