Skip to content

gpt-oss-safeguard-20b

ModelOpen sourceActive
by OpenAI · family “gpt-oss-safeguardb”

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... (description from the OpenRouter listing)

in: textout: textReasoningTool useStructured output
Epoch Capabilities Index
Not scored by Epoch AI
Listing ↗
Released
29 Oct 2025
Context window
131K tokens
Max output
66K tokens
Input $ / 1M tokens
$0.075
Output $ / 1M tokens
$0.30
Cached input $ / 1M
$0.037
Each benchmark shown separately with its own source

Benchmark results (0)

All benchmarks
No benchmark results recorded for this model.

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

12 recorded prices on 1 Nov 2025 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 3h ago. Steps show when the price changed.

Events