Urgent.News

What's breaking now, across thousands of outlets.

AI

Google Launches Gemini 3.8 Flash and Flash Cyber With New API Pricing and Access

Google has introduced Gemini 3.8 Flash and Gemini 3.8 Flash Cyber , two new models in its Flash family. Gemini 3.8 Flash targets coding, software engineering, agentic workflows and multi-step reasoning, while the Cyber variant is designed for vulnerability discovery and automated patching for trusted defenders. The release gives developers, paid Gemini users and Gemini Enterprise customers…

Google has unveiled Gemini 3.8 Flash and Flash Cyber, two new models in its Flash family, alongside updated API pricing and access. Gemini 3.8 Flash is tailored for coding, software engineering, agentic workflows, and multi-step reasoning, while the Flash Cyber variant is intended for vulnerability discovery and automated patching for trusted defenders.

Both models offer developers, paid Gemini users, and Gemini Enterprise customers varied access options. The Flash Flash models represent Google's third Flash release in six weeks, following Gemini 3.7 Flash. The company emphasizes that 3.8 Flash is its most intelligent workhorse model, delivering significant enhancements over 3.7 Flash on long-horizon reasoning and multi-step tasks.

For businesses, the new models are particularly valuable for real-world tasks that involve multiple steps, such as interpreting requests, working with code or documents, checking intermediate results, and producing outputs for review. Google positions Gemini 3.8 Flash around these coding and agent-oriented workflows, highlighting its acceleration through agentic loops and improvements in coding, software engineering, agentic workflows, long-horizon reasoning, and multi-step tasks.

Brief written by urgent.news from Dev.to's own syndicated text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

The Part Of Your RAG Pipeline That Decides Everything

Most teams debugging a disappointing RAG system start at the end of the pipeline. They swap the model, tune the prompt, raise top k.

  • Retrieval-augmented generation pipeline often fails due to ingestion stage issues.
  • Parsing non-text formats like PDFs is challenging and crucial for RAG quality.
  • Layout-aware parsers and metadata attachment are essential for effective chunking.

Everything I Didn't Build

For a few weeks I ran a proper multi-agent orchestrator. The concept was right, and I still think the people building those tools are pointed the correct way.

  • Author builds complex agent orchestrator with multiple components
  • System inefficient due to extensive communication overhead
  • Author simplifies using markdown, Git, and agent board

More from Friday 4 September →