gpt-oss-safeguard-20b
gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b.
openai.gpt-oss-safeguard-20bReleased 29 Oct 2025 (Oct 2025)Pricing
List prices at Amazon Bedrock, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $0.07 | $0.20 | |
| Batch 57% off | $0.03 | $0.10 | 57% off |
| Flex 57% off | $0.03 | $0.10 | 57% off |
| Priority 71% more | $0.12 | $0.35 | 71% more |
What it costs a month
At Amazon Bedrock list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out$36.00
- RAG over documents2,000 requests a day, 8K in / 0.6K out$40.80
- Coding agent1,000 requests a day, 40K in / 2K out$96.00
- Overnight batch20,000 requests a day, 4K in / 0.4K out, batch$96.00
Where to run it
3 hosts serve gpt-oss-safeguard-20b. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
Amazon BedrockCloud openai.gpt-oss-safeguard-20b | $0.07 / $0.20 | openai.gpt-oss-safeguard-20b | |
GroqInference gpt-oss-safeguard-20b | $0.075 / $0.30 +28% vs headline | gpt-oss-safeguard-20b | |
Fireworks AIInference gpt-oss-safeguard-20b | $0.50 / $0.50 +388% vs headline | gpt-oss-safeguard-20b |
Show 17 more listings: regions, snapshots and variantsHide regional listings
| Amazon Bedrock | |||||||
| Mantle openai.gpt-oss-safeguard-20b | $0.07 / $0.20 | openai.gpt-oss-safeguard-20b | |||||
| ap-southeast-2 · Mantle openai.gpt-oss-safeguard-20b | $0.0721 / $0.206 +3% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-south-1 openai.gpt-oss-safeguard-20b | $0.08 / $0.23 +15% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-west-1 openai.gpt-oss-safeguard-20b | $0.08 / $0.23 +15% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-south-1 · Mantle openai.gpt-oss-safeguard-20b | $0.08 / $0.23 +15% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-west-1 · Mantle openai.gpt-oss-safeguard-20b | $0.08 / $0.23 +15% vs headline | openai.gpt-oss-safeguard-20b | |||||
| ap-northeast-1 openai.gpt-oss-safeguard-20b | $0.08 / $0.24 +17% vs headline | openai.gpt-oss-safeguard-20b | |||||
| ap-south-1 openai.gpt-oss-safeguard-20b | $0.08 / $0.24 +17% vs headline | openai.gpt-oss-safeguard-20b | |||||
| sa-east-1 openai.gpt-oss-safeguard-20b | $0.08 / $0.24 +17% vs headline | openai.gpt-oss-safeguard-20b | |||||
| ap-northeast-1 · Mantle openai.gpt-oss-safeguard-20b | $0.08 / $0.24 +17% vs headline | openai.gpt-oss-safeguard-20b | |||||
| ap-south-1 · Mantle openai.gpt-oss-safeguard-20b | $0.08 / $0.24 +17% vs headline | openai.gpt-oss-safeguard-20b | |||||
| sa-east-1 · Mantle openai.gpt-oss-safeguard-20b | $0.08 / $0.24 +17% vs headline | openai.gpt-oss-safeguard-20b | |||||
| ap-southeast-3 · Mantle openai.gpt-oss-safeguard-20b | $0.084 / $0.24 +20% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-central-1 · Mantle openai.gpt-oss-safeguard-20b | $0.084 / $0.24 +20% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-north-1 · Mantle openai.gpt-oss-safeguard-20b | $0.084 / $0.24 +20% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-west-2 openai.gpt-oss-safeguard-20b | $0.11 / $0.31 +56% vs headline | openai.gpt-oss-safeguard-20b | |||||
| eu-west-2 · Mantle openai.gpt-oss-safeguard-20b | $0.11 / $0.31 +56% vs headline | openai.gpt-oss-safeguard-20b | |||||
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- Text
- Produces
- Text
- Open weights
- License: Apache-2.0Runs in about 51.6 GB of GPU memory
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than Amazon Bedrock.
Smaller OpenAI models
No cheaper OpenAI model keeps its key capabilities.
Other frontier labs
No cheaper match from another frontier lab.
Open-weight picks
No cheaper open-weight match.
Call it
Using Amazon Bedrock's API.
from litellm import completion
response = completion(
model="bedrock/openai.gpt-oss-safeguard-20b",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does gpt-oss-safeguard-20b cost?
On Amazon Bedrock, gpt-oss-safeguard-20b costs $0.07 per million input tokens and $0.20 per million output tokens.
Which APIs offer gpt-oss-safeguard-20b?
gpt-oss-safeguard-20b runs on 3 APIs: Amazon Bedrock, Groq and Fireworks AI.
What is gpt-oss-safeguard-20b's context window?
128K tokens, with up to 16K tokens in one reply.
Is gpt-oss-safeguard-20b being retired?
No retirement date is scheduled for it on any API we track.
When was gpt-oss-safeguard-20b released?
OpenAI released it on 29 Oct 2025.
What is gpt-oss-safeguard-20b's API model id?
Use openai.gpt-oss-safeguard-20b with Amazon Bedrock. Other hosts use their own ids: gpt-oss-safeguard-20b on Groq, gpt-oss-safeguard-20b on Fireworks AI.
Sources
Raw catalog entry for openai.gpt-oss-safeguard-20b on Amazon Bedrock
{
"input_cost_per_token": 7e-8,
"input_cost_per_token_batches": 3e-8,
"input_cost_per_token_flex": 3e-8,
"input_cost_per_token_priority": 1.2e-7,
"litellm_provider": "bedrock_converse",
"max_input_tokens": 128000,
"max_output_tokens": 16000,
"mode": "chat",
"output_cost_per_token": 2e-7,
"output_cost_per_token_batches": 1e-7,
"output_cost_per_token_flex": 1e-7,
"output_cost_per_token_priority": 3.5e-7,
"source": "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json",
"supports_audio_input": false,
"supports_bedrock_runtime_chat_completions_inline_reasoning": true,
"supports_function_calling": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_vision": false
}