OpenAI's gpt-oss-120b free tier is a notable offering: an intelligence index of 33.3 and a math index of 93.4 place it among the stronger open-access models available, and its livecodebench score of 0.878 is competitive with paid models well above its price point. Tool use and function calling are supported, making it usable for structured automation tasks.

The absence of vision and a missing agentic index limit its versatility compared to full-featured paid alternatives. The 131K context window is adequate for most business documents but falls short of the 400K available on the GPT-5 family. As a free model, it is best treated as a development and prototyping resource rather than a production backbone for mission-critical workflows.

For teams exploring OpenAI's open-weight direction or running cost-sensitive internal tooling, this model offers exceptional value. Its coding and math performance make it particularly useful for developer-facing applications, data analysis pipelines, and educational tooling where budget constraints are real.

Assessed June 30, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?