gpt-oss-safeguard-20b
OpenAI · Open weights · Released Oct 29, 2025 · openai/gpt-oss-safeguard-20b
Best $0.07 in / $0.20 out via Amazon Bedrock
131K contextToolsReasoningStructured outputPrompt cachingOpen weights
Available from 2 providersUSD / MTOK · PRICES 16 MINUTES AGO · SPEED 20 DAYS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | UPTIME |
|---|---|---|---|---|---|---|---|---|
| $0.07 | $0.20 | — | 131K | — | — | — | 100.0% | |
| $0.075 | $0.30 | $0.037 | 131K | — | 0.73s | 957 tok/s | 74.1% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · methodology v1.1 · measured 20 days ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Recent changesRSS ↗
Nov 20, 2026endpoint added · amazon-bedrock1 → 2
Nov 5, 2026endpoint removed · cheaperinference2 → 1
Oct 23, 2026endpoint added · cheaperinference1 → 2
About
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...
Specifications
Context window131,072
Max output65,536
Modalities intext
Modalities outtext
LicenseOpen weights
ReleasedOct 29, 2025
TokenizerGPT
WeightsHugging Face ↗
Open weights
safetensors
Related
Embed badge
[](https://modelindex.ai/models/openai/gpt-oss-safeguard-20b)Sources
OpenRouter API16 minutes ago
AGGREGATORHugging Face16 minutes ago
COMMUNITYModelIndex benchmarks20 days ago
MEASUREDVerified 16 minutes ago · list prices, not negotiated rates.