Genkit

Bring your old photos back to life with Genkit Go and Gemini! 📸✨ Check out this excellent tutorial by Daniela Petruzalek, demonstrating how to use Genkit with Nano Banana Pro (Gemini 3 Pro Image) to restore, colorize, and upscale vintage photos with high fidelity. It's a fantastic showcase of how simple building powerful AI applications with Go and Gemini can be! Want to learn how to build it for yourself? Check out the codelab: https://lnkd.in/eKwEVNAb

Bring your old photos back to life with Genkit Go and Gemini! 📸✨ Check out this excellent tutorial by Daniela Petruzalek, demonstrating how to use Genkit with Nano Banana Pro (Gemini 3 Pro Image) to restore, colorize, and upscale vintage photos with high fidelity. It's a fantastic showcase of how simple building powerful AI applications with Go and Gemini can be! Want to learn how to build it for yourself? Check out the codelab: https://lnkd.in/eKwEVNAb

🚀 Genkit JS & CLI v1.29.0 is officially live! We shipped major updates to models, core APIs, and developer tooling to supercharge your AI workflows. Here’s what’s new: ✨ Gemini 3.1 & Veo 3.1 support via the Google GenAI plugin ✨ Improved Anthropic prompt caching support to reduce latency & costs ✨ A new `cancelOperation` API for cleanly managing background models ✨ Agentic docs discovery in the CLI & MCP to empower your coding agents ✨ OpenAI `gpt-4o-mini-transcribe` support A huge shoutout to our new community contributors for this release: 7hokerz, traut, prashant4nov, dmeremyanin, and danicat! 🙌 Full release notes: https://lnkd.in/e4WjjTzP

🚀 Amazon Bedrock and Azure AI Foundry are now officially supported in Genkit Go and Genkit JS Build once, run anywhere, cloud-native. Amazon Bedrock ⬇️ Go: https://lnkd.in/gKQBmbsU JS: https://lnkd.in/g5rjbTET Azure ⬇️ Go: https://lnkd.in/gCh7M_W5 JS: https://lnkd.in/gDWnZHFX Thank you to Xavier Portilla Edo, Genkit's top community contributor, for building and contributing these plugins!

Once you've defined a flow, securely exposing it over HTTP is straightforward. Genkit handles JSON serialization, and streaming flows automatically use Server-Sent Events. Define your AI logic, expose it over HTTP. Streaming flows get SSE automatically. Sample: https://lnkd.in/g7wt72Q7

📊 Prompt Evaluations have officially arrived in the Genkit Dev UI! With the release of Genkit 1.28.0, you can now move beyond "vibes" and actually measure your prompt performance: 🧪 Run & Score: Bulk run prompts against datasets and score them using prepackaged or custom metrics. 🎛️ Live Tuning: Adjust model configuration on the fly. 👯♂️ Side-by-Side Comparison: Compare different prompt versions or configs to find your winner. Level up your prompt engineering today. 🚀 Learn more about Genkit Evaluation features: https://lnkd.in/gZxyKXYH

We just released **Genkit JS 1.28.0**! 🚀 This update packs some powerful features for building AI applications: 🔹 Anthropic Integration: The `@genkit-ai/anthropic` plugin is now officially supported! It supports Structured Output and Citations, giving you more control and reliability with Claude models. Check out the docs: https://lnkd.in/g3AjzMXi 🔹 Prompt Evaluation: We've added new features to help you evaluate your prompts and model responses effectively. Learn more: https://lnkd.in/gZxyKXYH 🔹 Platform Support: Important fixes to unblock running Genkit on Cloudflare Workers. Docs: https://lnkd.in/gjqubQFd 🔹 Antigravity: Fixed MCP server startup issues. The Genkit MCP server is now available in the Antigravity MCP Store. 🔹 Vertex AI: Added support for the Virtual Try On model. Read the full release notes: https://lnkd.in/gg5f7KwH

Did you know you can deploy Genkit flows as AWS Lambdas and integrate AWS models (Bedrock)? And it doesn’t stop there; you can develop and test your Lambda locally while using the Genkit UI. I break it all down step by step here:

If you're using Go for AI, you probably want type safety on both sides of the prompt — the input and the output. DataPrompt lets you define typed inputs and get typed outputs. Your IDE autocompletes, your compiler catches mistakes. Type safety from input to output — that's how prompts should work in Go. Check out the docs to get started: https://lnkd.in/g8-87xdt

A solid deep dive on AI development for Go devs. 👇 The sections on Flows (at 5:32) and Observability (at 11:42) really show off the power of Genkit. Excellent walkthrough by Daniela Petruzalek. #Golang #GenAI #Genkit