GPT-OSS Safeguard 20B
JSON →A 20-billion-parameter safety classifier from OpenAI designed to detect harmful content and enforce usage policies in open-source deployments.
Specs
context window 131K tokens
max output 66K tokens
input price $0.075 / 1M tokens
output price $0.3 / 1M tokens
Capabilities
json-modestreamingopen-weightsfunction-callingtool-usereasoning
Dates
releasedMar 2025
knowledge cutoffJun 2024
Resources
API
full doc /v1/models/gpt-oss-safeguard-20b