Oracle is adding Google’s Gemini models to the Oracle Cloud Infrastructure (OCI) Generative AI service, creating new opportunities for partners who deliver managed services and AI-driven solutions. The initial rollout includes Gemini 2.5, giving enterprises the ability to build agents that support multimodal understanding, advanced software development, workflow automation, and research.
Broadening Model Access
Oracle isn’t stopping with a single model. The company plans to integrate the full range of Gemini models into OCI through Vertex AI, extending access to advanced capabilities for video, image, speech, and music generation. Industry-specific models like MedLM for healthcare are also in scope. Over time, Gemini options will appear inside Oracle Fusion Cloud Applications, enabling AI-driven enhancements across finance, HR, supply chain, sales, and service. By using existing Oracle Universal Credits, customers can activate Gemini quickly, without renegotiating contracts or adding separate licensing costs.
Implications for Partners
For partners, the combination of Oracle workloads and Google Cloud AI creates a direct path to delivering higher-value services.
As Google's spokesperson explained, "Channel partners are very important to helping customers combine Oracle workloads with Google Cloud’s AI technology. Now, it’s even easier for these partners to access Google’s powerful Gemini models within products and services that Oracle customers may already use, such as the OCI Generative AI service via new Vertex AI integrations.”
This means MSPs, MSSPs, and VARs can bring AI-powered capabilities into environments their clients already trust, instead of introducing standalone tools that may complicate adoption. It also positions partners to design sector-specific offerings, for example, healthcare use cases with MedLM or finance workflows enhanced by embedded Gemini models, while staying within the Oracle ecosystem customers are already invested in.
The Infrastructure Layer
Behind the model access is OCI’s focus on infrastructure. Bare metal GPU instances are optimized for demanding AI workloads, from generative models to natural language processing and computer vision. For partners, having both the models and the compute resources in one environment simplifies deployment, reduces integration friction, and creates more predictable performance at scale.
By tying Google’s Gemini models directly into Oracle’s cloud services, this collaboration gives partners a new way to expand their service portfolios, whether that’s delivering AI agents for workflow automation, embedding AI into enterprise applications, or running complex AI workloads for regulated industries
