Open Models Launch Initiated by Groq and HUMAIN on Day Zero, Overseen by OpenAI
GroqCloud, a leading AI inference platform, has announced the immediate availability of OpenAI's gpt-oss-120B and gpt-oss-20B models globally. This launch marks a significant step forward in making cutting-edge AI technology accessible to developers and enterprises around the world.
Global Reach and Minimal Latency
Through GroqCloud, OpenAI's open models are now available worldwide with minimal latency. The collaboration with HUMAIN, Saudi Arabia's leading AI services provider and a PIF company, strengthens local enterprise access and support, particularly in the Middle East. This ensures that public and private sector organizations can deploy these models with regional optimization and support.
Groq's global data center footprint, including locations in North America, Europe, and the Middle East, guarantees reliable, high-performance AI inference worldwide.
Pricing and Performance
The pricing for OpenAI's open models on GroqCloud is competitive, offering affordable inference for developers. For a limited time, tool calls used with these models will not be charged.
- The pricing for gpt-oss-120B on GroqCloud is $0.15 per million (M) input tokens and $0.75 per M output tokens.
- The pricing for gpt-oss-20B on GroqCloud is $0.10 per M input tokens and $0.50 per M output tokens.
Groq delivers extended context and built-in tools like code execution and web search. The models are available with full 128K context, real-time responses, and integrated server-side tools.
High Throughput and Open-source Openness
The high throughput of these models is impressive. gpt-oss-120B handles 500+ tokens per second, while gpt-oss-20B reaches 1000+ tokens per second on GroqCloud.
Open-source openness is another key benefit. The models are released under Apache 2.0 and freely available on Hugging Face, allowing customization and fine-tuning for specific use cases.
Developer-friendly Access
GroqCloud provides a Developer Tier with self-serve, pay-as-you-go access and increased rate limits, plus a Batch API for cost efficiency and scalable workloads. This makes it easier than ever for developers to access and utilise these powerful models.
In summary, GroqCloud offers an affordable, high-performance, and globally accessible deployment of OpenAI’s gpt-oss-120B and gpt-oss-20B models, coupled with local support in Saudi Arabia and a flexible pricing structure geared for developers and enterprises worldwide.
- The collaboration between GroqCloud and HUMAIN, a leading AI services provider in Saudi Arabia, offers increased access and support for local enterprises in the Middle East, enabling them to deploy OpenAI's models with regional optimization and support.
- With its competitive pricing and extended context, GroqCloud empowers developers by providing affordable inference for the OpenAI models, offering self-serve, pay-as-you-go access, increased rate limits, a Batch API, and built-in tools like code execution and web search.
