Google Deepmind rolls out Gemini 2.0 Flash for general use, introduces Pro experimental for coding, and unveils Flash-Lite for cost-efficient AI development

Google DeepMind has officially launched the latest versions of its Gemini 2.0 AI models, making them widely accessible to developers and users. The rollout includes updates to Gemini 2.0 Flash, now generally available via the Gemini API in Google AI Studio and Vertex AI, as well as the debut of Gemini 2.0 Pro Experimental, designed for superior coding performance and handling complex prompts.

Gemini 2.0 Flash, first introduced at I/O 2024, has become a favourite among developers due to its high-speed processing, multimodal capabilities, and extensive 1-million-token context window. The model is now fully available across Google’s AI platforms, with upcoming support for image generation and text-to-speech.

Additionally, Google has introduced Gemini 2.0 Flash-Lite, a cost-efficient AI model now in public preview, optimised for large-scale, high-frequency tasks.

The highlight of the release is Gemini 2.0 Pro Experimental, Google’s most advanced model yet for coding, complex reasoning, and vast data analysis. With an expanded 2-million-token context window, it offers enhanced world knowledge, code execution, and tool integration, including Google Search support.

Google plans to expand Gemini 2.0’s multimodal capabilities in the coming months, with additional functionalities expected soon. Developers and AI enthusiasts can explore these models via the Gemini API, Google AI Studio, and Vertex AI. Full details, including pricing and benchmarks, are available on the Google for Developers blog.