AWS / Model configuration
gpt-oss-120b · openai.gpt-oss-120b-1:0
AU processing documentedOpenAI’s open-weight 120B model hosted by Bedrock in-region in Sydney. This is not the OpenAI API.
Source review: 09 Nov 2026 · Named model configuration; confirm live account availability
Model coverage
gpt-oss-120b · openai.gpt-oss-120b-1:0
API access
AWS account; model access and quota required
Billing
Standard on-demand; USD $0.1545 input / $0.618 output per 1M tokens
Capacity commitment
No reserved capacity required for this service category.
Storage
Prompts and outputs are processed in the selected AWS region. Bedrock retention is configured separately and varies by model.
Retention and training
Check model-specific abuse-detection requirements. Zero retention is configurable for supported models and is not assumed by this entry.
Ownership and control
AWS operates Bedrock infrastructure; OpenAI (open-weight release) is the model developer.
What to check before using it
- Call ap-southeast-2 directly using openai.gpt-oss-120b-1:0.
- Confirm the model is enabled for the account and check quota.
- Review Bedrock retention settings and the model provider's terms.
Still to verify
- Account-specific enablement and rate limits
- Model-specific abuse-detection retention
- Independent operational test not performed
Sources and supported claims
Model support by AWS Region ↗
The Bedrock model support matrix lists gpt-oss-120b as In-Region in ap-southeast-2 (Sydney), with Geo and Global profiles not supported.
Checked 2026-11-09 · Scope: this model configurationAmazon Bedrock pricing ↗
AWS Price List data for ap-southeast-2 (published 6 Nov 2026) lists gpt-oss-120b Standard at USD $0.1545 input / $0.618 output per 1M tokens.
Checked 2026-11-09 · Scope: this model configurationData retention controls ↗
Bedrock data retention is a separate account/model control; do not assume zero retention.
Checked 2026-11-09 · Scope: this model configuration