GPT-OSS Now Live on Core42 AI Cloud via Compass

Core42 Compass gives you choice with no lock in. Deploy sovereign AI solutions at unmatched speeds on the silicon that suits your workloads.

Get Started Now

Unmatched Performance - Hit world-record inference speeds – an incredible 3,000 tokens/second/user on Cerebras clusters - for real time AI at scale.
Silicon Freedom - Choose from Nvidia, Microsoft Azure or Cerebras. Align each workload with the optimal accelerator.
Sovereign AI - All UAE hosted deployments run under UAE jurisdiction. Benefit from locally optimized infrastructure with global-class performance and full regulatory compliance.
Instant Access - Live today through Core42 Compass API. Flexible pricing, custom integrations and enterprise-grade support from day one.

Competitive Pricing for gpt-oss Models

Simple, transparent pricing across three silicon platforms that scales with your usage

Model/Deployment

Silicon

Input $/M

Output $/M

gpt-oss-120b

Cerebras

0.25

0.69

gpt-oss-120b

NVIDIA HGX

0.30

0.75

gpt-oss-120b Microsoft

Microsoft Azure

0.40

0.95

gpt-oss-20b-core42

NVIDIA HGX

0.20

0.60

Set custom rate limits that work for you: adjust requests per minute, token volumes, and throughput quotas to match your application's unique demands

Build with gpt-oss-120B & gpt-oss-20B in Minutes!

Ready to Get Started?

Get Started Now

Talk to Sales

Our team can tailor a plan to meet your needs.

Speak to an Expert