OpenAI: gpt-oss-safeguard-20b

OpenAIID: openai/gpt-oss-safeguard-20b

gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust & safety labeling. Learn more about this model in OpenAI's gpt-oss-safeguard [user guide](https://cookbook.openai.com/articles/gpt-oss-safeguard-guide).

Pricing per 1M Tokens

Input (Prompt)$0.07
Output (Completion)$0.30
Cache Read$0.04
Cache WriteFree
ImageN/A

Specifications

Context Length131K
Max Output Tokens66K
Input ModalitiesText
Output ModalitiesText
TokenizerGPT
Instruct TypeN/A
Top Provider Context131K
Top Provider Max Output66K
ModeratedNo

Compare this model

See how OpenAI: gpt-oss-safeguard-20b stacks up against other models.

vs OpenAI: GPT-5.4 Nanovs OpenAI: GPT-5.4 Minivs OpenAI: GPT-5.4 Provs OpenAI: GPT-5.4vs OpenAI: GPT-5.3 Chat

More from OpenAI

OpenAI: GPT-5.4 Nano
Input$0.20
Context400K
OpenAI: GPT-5.4 Mini
Input$0.75
Context400K
OpenAI: GPT-5.4 Pro
Input$30
Context1.1M
OpenAI: GPT-5.4
Input$2.50
Context1.1M
OpenAI: GPT-5.3 Chat
Input$1.75
Context128K

Last updated: March 23, 2026

First tracked: March 23, 2026