Unlocking Potential: Google’s Advanced AI Gemini Now Ready for Enterprise Development

Google has officially launched its most advanced generative AI model, Gemini, for enterprise app development needs.

Unveiled last week, Gemini is available in three versions: Ultra, Pro, and Nano. Today's announcement makes the Pro version accessible through an API, allowing developers to use it for free with certain usage limits, as detailed in a recent blog post.

Gemini Pro for Developers: Key Features

Developers can access the first version of Gemini Pro via the Google AI Studio API, providing a web-based platform for creating prompts and obtaining API keys for app development. This version offers a 32K context window for text generation, with plans to expand this feature in the future.

Google has also introduced a dedicated Gemini Pro Vision multimodal endpoint that accepts both text and image inputs, providing text outputs. In a post on X, CEO Sundar Pichai highlighted the comprehensive features of the Gemini API, including function calling, embeddings, semantic retrieval, custom knowledge grounding, and chat capabilities. The API supports 38 languages across more than 180 countries.

Gemini Pro will also be integrated into Vertex AI, Google Cloud's end-to-end AI platform, which includes tools, fully-managed infrastructure, and built-in privacy and safety features. This integration allows developers to transition to a managed environment as needed.

The company aims to gather feedback from developers to refine Gemini Pro as it moves toward launching the more complex Gemini Ultra next year.

Free Access with Limitations

Currently, Google offers Gemini Pro and Gemini Pro Vision for free, with a rate limit of 60 requests per minute. This applies to developers using the models on Vertex AI as well, but this free access is available only until general availability next year. Notably, Google’s free quota is 20 times larger than offerings from competitors, making it suitable for most development projects.

Once the service is fully available, pricing will be implemented based on usage, charging per 1,000 characters or images. Specifically, the input price for Gemini Pro is set at $0.00025 per 1K characters and $0.0025 per image, while the output remains at $0.0005 per 1K characters.

Some users on X have noted that Google’s pricing model, which charges per character, is significantly higher than competitors like OpenAI, which typically charges per token—a numeric representation that can encompass entire words.

Enhancements in Vertex AI

Alongside the Gemini Pro launch, Google has updated Vertex AI with its latest text-to-image diffusion technology, Imagen 2. This upgrade introduces features to generate a broad range of creative and realistic logos, emblems, and lettermarks, while also improving results in rendering text across multiple languages.

Additionally, Google announced the availability of MedLM, a family of foundation models fine-tuned for the healthcare sector, to US-based organizations through Vertex AI. This new offering builds on the earlier Med-PaLM 2 foundation model, with a Gemini-based upgrade expected soon.