Name: OpenAI: gpt-oss-safeguard-20b
Author: openai

Question 1

What is gpt-oss-safeguard-20b?

Accepted Answer

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(opens in new tab).

Question 2

How much does gpt-oss-safeguard-20b cost?

Accepted Answer

gpt-oss-safeguard-20b costs $0.075/M input tokens and $0.30/M output tokens, with separate rates for Cache Read at $0.0375/M tokens.

Question 3

What is the context length of gpt-oss-safeguard-20b?

Accepted Answer

gpt-oss-safeguard-20b has a 131,072 token context window. It supports up to 65,536 completion tokens.

Question 4

Does gpt-oss-safeguard-20b support tool calling and structured outputs?

Accepted Answer

Yes. gpt-oss-safeguard-20b accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Question 5

What other text models does OpenAI have?

Accepted Answer

GPT-5.6 Luna Pro, GPT-5.6 Luna, GPT-5.6 Terra Pro and 55 more are other text models from OpenAI.

Question 6

When was gpt-oss-safeguard-20b released?

Accepted Answer

gpt-oss-safeguard-20b was released on 2025-10-29.

OpenAI: gpt-oss-safeguard-20b

openai/gpt-oss-safeguard-20b

OpenAI: gpt-oss-safeguard-20b

openai/gpt-oss-safeguard-20b

Frequently asked questions

Frequently asked questions