Skip to content

Listings

All listings for GPT OSS Safeguard 20B

Every priced listing we track for this model. Back to GPT OSS Safeguard 20B

ProviderMax Input TokensInput $/1MOutput $/1MTok/sLatency
Run by the model's maker
 OpenAI
gpt-oss-safeguard-20bSource: GenAI Prices data
Data from GenAI Prices
Not published by the provider$0.075cache read $0.037$0.30——
 Amazon Web Servicesopenai.gpt-oss-safeguard-20bSources: AWS Bedrock Runtime data, LiteLLM data, Models.dev data
Data from AWS Bedrock Runtime
Fields by sourceFrom LiteLLM: offering name, context length, max input tokens, max output tokens, supported endpoints, supports function calling, supports tool choice, supports response schema, supports batching, supports reasoning, max input tokens, max output tokens, currency, last updated, input tokens, output tokens, input tokens batch, output tokens batchFrom Models.dev: max completion tokens, supports system messages, supports vision, supports image input
131K$0.07batch $0.03$0.20batch $0.10——
 Eden AIInference by Amazon Web Servicesamazon/openai.gpt-oss-safeguard-20bSource: Models.dev data
Data from Models.dev
128K$0.07$0.20——
 Eden AIInference by Amazon Web ServicesRegions: 🇺🇸 usamazon/openai.gpt-oss-safeguard-20b@usSource: Models.dev data
Data from Models.dev
128K$0.07$0.20——
 Eden AIInference by Groqgroq/openai/gpt-oss-safeguard-20bSource: Models.dev data
Data from Models.dev
131K$0.075cache read $0.0375$0.30——
 Hugging FaceInference by Amazon Web Servicesopenai/gpt-oss-safeguard-20bSource: Hugging Face Inference Providers data
Data from Hugging Face Inference Providers
131K$0.07$0.20——
 Kiloopenai/gpt-oss-safeguard-20bSources: Kilo AI Gateway data, Models.dev data
Data from Kilo AI Gateway
Fields by sourceFrom Models.dev: supports system messages, supports vision, max input tokens, max output tokens, supports image input
131K$0.07$0.20——
 NanoGPT
zero retention, per NanoGPT
Data privacy: zero data retention · prompts not retained · no training on prompts Source: NanoGPT
policy notes
privacy: zdr; distillation: License permits distillation — NanoGPT privacy policy
Inference by KitaniRegions: GLBopenai/gpt-oss-safeguard-20b · mxfp4Source: NanoGPT
Data from NanoGPT
Not published by the provider$0.0709cache read $0.0354$0.294—32,926ms
 NanoGPTopenai/gpt-oss-safeguard-20bSources: Models.dev data, NanoGPT
Data from NanoGPT
Fields by sourceFrom Models.dev: max completion tokens, supports reasoning, supports system messages, supports vision, max input tokens, max output tokens, supports image input
131K$0.075$0.30——
 NanoGPT
zero retention, per NanoGPT
Data privacy: zero data retention · prompts not retained · no training on prompts Source: NanoGPT
policy notes
privacy: zdr; distillation: Provider terms restrict distillation — NanoGPT privacy policy
Inference by GroqRegions: 🇺🇸 USopenai/gpt-oss-safeguard-20bSource: NanoGPT
Data from NanoGPT
Not published by the provider$0.0788cache read $0.0394$0.315—306ms
 Nous Portalopenai/gpt-oss-safeguard-20bSource: Nous Portal Inference Gateway data
Data from Nous Portal Inference Gateway
131K$0.075cache read $0.0375$0.30——
 OpenRouterInference by Amazon Web Servicesopenai/gpt-oss-safeguard-20b · us-east-2Source: OpenRouter data docs
Data from OpenRouter
131K$0.07$0.20233.5362.5ms
 OpenRouteropenai/gpt-oss-safeguard-20bSources: GenAI Prices data, Models.dev data, OpenRouter data docs
Data from OpenRouter
Fields by sourceFrom GenAI Prices: offering nameFrom Models.dev: supports system messages, supports vision, max input tokens, max output tokens, supports image input
131K$0.075cache read $0.0375$0.30——
 Vercel
policy notes
AI Gateway uses zero data retention by default — Vercel AI Gateway security & compliance documentation It permanently deletes your prompts and responses after requests complete. — Vercel AI Gateway security & compliance documentation requests only route to providers with verified ZDR agreements — Vercel AI Gateway security & compliance documentation The provider allowlist lets team owners restrict which AI providers can serve requests through AI Gateway — Vercel AI Gateway security & compliance documentation Vercel may store, process and transmit your information outside of your country of residence, which may have different data protection laws and may not offer the same level of protection or guarantees as in your country or the country where you initially provided the information. — Vercel privacy policy Sources: Vercel AI Gateway security & compliance documentation, Vercel privacy policy
openai/gpt-oss-safeguard-20bSources: Models.dev data, Vercel AI Gateway data
Data from Vercel AI Gateway
Fields by sourceFrom Models.dev: max completion tokens, supported output modalities, supports function calling, supports system messages, supports vision, max input tokens, max output tokens, supports image input
128K$0.07$0.20——
 Aimlapiopenai/gpt-oss-safeguard-20bSource: Models.dev data
Data from Models.dev
131K$0.0794cache read $0.0397$0.3174——
 Fireworks AIfireworks_ai/accounts/fireworks/models/gpt-oss-safeguard-20bSource: LiteLLM data
Data from LiteLLM
131K$0.50$0.50——
 Groqopenai/gpt-oss-safeguard-20bSources: Groq API data, Groq documentation, LiteLLM data, Models.dev data
Data from Groq API
Fields by sourceFrom LiteLLM: offering name, max input tokens, max output tokens, supported input modalities, supported output modalities, supports function calling, supports tool choice, supports response schema, supports prompt caching, supports reasoning, supports web search, max input tokens, max output tokens, currency, last updated, input tokens, output tokens, cache-read priceFrom Models.dev: max completion tokens, supports system messages, supports vision, supports image inputFrom Groq documentation: is available
131K$0.075cache read $0.0375$0.30——
 Lowrouterauto/openai/gpt-oss-safeguard-20bSource: Models.dev data
Data from Models.dev
131K$0.08$0.24——
 Merge Gatewayopenai/gpt-oss-safeguard-20bSource: Models.dev data
Data from Models.dev
4K$0.07$0.20——

4 listings without a price are hidden.

Select any dotted value, label or icon to see where it comes from and how it is measured. How this data works