Google for Developers

Gemma 4 is here! Our most intelligent open models to date, are built on the same world-class research and tech as Gemini 3, and are sized to run and fine-tune efficiently on local hardware. Check out what @GoogleGemma 4 brings to devs: 💎 Advanced Reasoning: Deep logic tasks, complex multi-step planning, and beyond 💎 Longer context: Seamlessly analyze entire codebases with context windows of 128K tokens for our edge models and 256K tokens for our largest models 💎 Vision and audio: Rich, multimodal interactions out of the box 💎 140+ languages: Trained on 140+ languages 💎 Apache 2.0 license: industry-standard open-source license

✅ Good caching = fast sites. ❌ Bad caching = fast exits.

@Vanarchain Couldn't agree more. ✨

@googledevs How about all? 👀

When building, what task do you find AI the most helpful for?

@sarahyang_ai Happy experimenting!

@googledevs @googlegemma thinking about how much easier it’ll be to actually experiment when you don’t need to worry about cloud costs

Gemma 4 is here! Our most intelligent open models to date, are built on the same world-class research and tech as Gemini 3, and are sized to run and fine-tune efficiently on local hardware. Check out what @GoogleGemma 4 brings to devs: 💎 Advanced Reasoning: Deep logic tasks, complex multi-step planning, and beyond 💎 Longer context: Seamlessly analyze entire codebases with context windows of 128K tokens for our edge models and 256K tokens for our largest models 💎 Vision and audio: Rich, multimodal interactions out of the box 💎 140+ languages: Trained on 140+ languages 💎 Apache 2.0 license: industry-standard open-source license

@googleaidevs @googlegemma The open source Apache 2.0 license is huge! Giving the dev community complete control over data, infrastructure, and models. So excited to see what the community creates! 🎉

Welcome to the family, @GoogleGemma 4️⃣ Gemma 4 are our most intelligent open models designed to run efficiently on every device, engineered to give developers total control over their deployments.

Explore the latest dev updates to help you build and scale 🧵✨

Balance cost & reliability with our new Flex & Priority inference tiers in the Gemini API! Flex: Pay 50% less for cost-sensitive & latency-tolerant workloads Priority: Highest reliability for your most critical, interactive apps (with premium pricing) Together with the async Batch API, these synchronous tiers give you a complete set of options for any workload. Just swap the tier with a single line of code and keep building.

See what Gemma 4 can do for your Android app!

Go beyond chatbots. Build your next AI agent on your own device. 🤖 Use #GoogleAIEdge to bring Gemma 4's power to the edge. Experiment in the Gallery app or deploy to any device – laptop, mobile, IoT – via LiteRT-LM. What will you build? Learn how: goo.gle/4tpurAO

@googlegemma Start building with the Gemma 4 family: goo.gle/3NTSiJV

@LeoTava8 Perfectly said.

@googledevs Progressive disclosure for domain expertise is the right move. Keeps the core agent lean and avoids context window bloat until you actually hit a specialized boundary. Separation of concerns for agents is underrated.

Build more efficient AI agents with the Agent Skills specification 🛠️ By using progressive disclosure, you can load domain expertise only when needed. This can reduce baseline context usage by 90%. We break agent knowledge into three layers: 1️⃣ L1 metadata: Just enough info for relevance 2️⃣ L2 instructions: Full skill body on demand 3️⃣ L3 resources: External files pulled when required Explore the full guide by @lavinigam and @Saboo_Shubham_: goo.gle/4bWZ7Cm

@SvenUrbanSci Love this take.

@googledevs The meta-skill factory pattern is the architectural breakthrough here. Enabling agents to generate and load their own SKILL.md files at runtime creates truly self-extending capabilities.

@golang Read the blog: goo.gle/41G4bpS

@golang ADK Go samples: goo.gle/4bT6Ktu

Build production-grade AI agents with @golang. ADK for Go 1.0 is here ⭐ Check out the new capabilities for building robust and debuggable agentic systems: 🔎 Native OpenTelemetry integration 🧩 Plugin system for extensibility 🛡️ HITL confirmation for enhanced security ⚙️ YAML-based agent configuration 🔗 Refined A2A protocol