OpenAI: gpt-oss-120b (free)
The free tier of gpt-oss-120b offers strong math and livecodebench scores at no cost, with solid instruction-following capability — an excellent option for teams wanting to evaluate OpenAI's open-weight model without upfront spend.
Assessment date: March 12, 2026
Our methodology takes into account a range of factors including pricing, functionality, capabilities, benchmark performance, and real-world applicability. Rankings are reviewed and updated regularly as new models are released. Issues with our rankings? Contact us
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.
Capabilities
Architecture
| Modality | Text → Text |
| Tokenizer | GPT |
| Parameters | 120B |
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Evaluations
Benchmark data from Artificial Analysis and Hugging Face
Model Information
Live Performance
Live endpoint metrics — refreshed every 30 minutes.
