All news with #vertex ai tag

🖼️ Google announced native image generation and editing in Gemini 2.5 Flash Image, now available in preview on Vertex AI. The model delivers state-of-the-art capabilities including multi-image fusion, character and style consistency, and conversational editing to refine visuals via natural-language loops. Built-in SynthID watermarking supports responsible, transparent use. Developers and partners report promising integrations and low-latency performance for real-time editing workflows.

⚙️ This guide from Google Cloud authors Eric Hanley and Brittany Rockwell explains how to tune vLLM deployments for xPU inference, covering accelerator selection, memory sizing, configuration, and benchmarking. It shows how to gather workload parameters, estimate HBM/VRAM needs (example: gemma-3-27b-it ≈57 GB), and run vLLM’s auto_tune to find optimal gpu_memory_utilization and throughput. The post compares GPU and TPU options and includes practical troubleshooting tips, cost analyses, and resources to reproduce benchmarks and HBM calculations.