gpt-oss-120b&gpt-oss-20b Model Card

By OpenAI Sandhini Agarwal, L. Ahmad, Jason Ai, S. Altman, Andy Applebaum, Edwin Arbus, Rahul K. Arora, Yu Bai, Bowen Baker, Hai-Biao Bao and 114 more. 2025

1,045 citations recorded by the source corpus.

We present gpt-oss-120b and gpt-oss-20b, two open-weight reasoning models that push the frontier of accuracy and inference cost. The models use an efficient mixture-of-expert transformer architecture and are trained using large-scale distillation and reinforcement learning. We optimize the models to have strong agentic capabilities (deep research browsing, python tool use, and support for developer-provided functions), all while using a rendered chat format that enables clear instruction following and role delineation. Both models achieve strong results on benchmarks ranging from mathematics, coding, and safety. We release the model weights, inference implementations, tool environments, and tokenizers under an Apache 2.0 license to enable broad use and further research.

View Paper: https://www.semanticscholar.org/paper/2a609a89709d353d0b67be2babaafb5ee9e1188a

Citations in the corpus

• TrajPrism: A Multi-Task Benchmark for Language-Grounded Urban Trajectory Understanding (2026, unknown): “To balance quality and scale, we adopt a two-stage generation strategy: approximately 2K few-shot seed examples are first produced using Gemini 3.1 Flash-Lite 7 and Claude Sonnet 4.6 8 , then used as in-context demonstrations for large-scale generation with GPT-OSS 120B [1] across all three cities.”
• ER-MIA: Black-Box Adversarial Memory Injection Attacks on Long-Term Memory-Augmented Large Language Models (2026, unknown): “2:3b as the victim LLM, with additional experiments on Gemma 3:27b (Team et al., 2025) and GPT-oss:20b (Agarwal et al., 2025).”
• SHERLOC: Structured Diagnostic Localization for Code Repair Agents (2026, unknown): “The quality-filtered row should therefore be read as an analysis-time selection experiment; substituting an open-weight judge (e.g., GPT-OSS [OpenAI, 2025b]) for GPT-5 .”