Google Open Source

Announcing new open source releases, exploring projects, sharing how we approach FOSS, and supporting communities around the world.

[Google Open Source @GoogleOSS]

Open source is free as in puppies, not free as in beer. 🐶📦

While prior research focuses on upstream maintenance, @sophia_iv studied the effort required to USE and update open source packages at Google.

👇 Link in first reply

[Google Open Source @GoogleOSS]

@omt66 *stares in Gemma 4, LiteRT-LM, gemma.cpp, and llm-d* 👀

More open stacks help everyone. Google teams ship Apache 2.0 inference from upstream vLLM & Kubernetes serving to local devices:
https://t.co/EIh9GzykOX

[Google Open Source @GoogleOSS]

@AgenticAIFdn Heading into #AGNTCon + #MCPCon NA this week? Read the full maintainer roundtable recap with Google's Kuba Herczyński alongside Anthropic, https://t.co/160zQpxlEU, and OpenAI: https://t.co/cjAm1iiZDr

[Google Open Source @GoogleOSS]

Open standards work when peers compare notes in neutral spaces.

At @AgenticAIFdn's EU roundtable, maintainers from Google, Anthropic, https://t.co/160zQpxlEU, and OpenAI unpacked MCP + A2A, narrow agent identity, and task cost.

👇 Link in first reply

[Google Open Source @GoogleOSS]

https://t.co/4eg4rL6S6b #PyTorch

[Google Open Source @GoogleOSS]

Join Sandeep Pokkunuri & Chris Jones on Oct 20 at 11:45 AM for "Cloud TPU Nexus: Autonomous Multi-Agent Swarms for PyTorch"!

Learn to streamline PyTorch-to-TPU migrations and achieve >70% of hand-tuned performance in <24 hours.

[Google Open Source @GoogleOSS]

🗓️ Add it to your schedule https://t.co/4eg4rL6S6b

[Google Open Source @GoogleOSS]

Join Rob Mulla at 10:40 AM for our sponsored demo: Unifying & Accelerating Open-Source LLM Serving on Google Cloud TPUs with TorchTPU!

See how it provides a unified backend for vLLM & SGLang, deploying SOTA models with minimal code changes.

[Google Open Source @GoogleOSS]

PyTorch Conference 2026 is live! We’re kicking off 2 days of open source innovation with @PyTorch. Stop by booth D5 to chat with our team & grab some swag! 🌐✨ Let us know if you're here! #PyTorchCon

[Massimo Asuni @asunimassimo]

New on the @GoogleOSS blog: This Week in Open Source covers 35 years of Linux, the 2026 State of OSPOs in the age of AI, terminal A2A/MCP agent workflows, and local multimodal search:
https://t.co/bKPWrwTDNQ #OpenSource https://t.co/azvSwqErQs

[Google Open Source @GoogleOSS]

Stop by the Google Community Lounge at #AllThingsOpen! ☕

It’s the place to recharge, network, and chat with members of the @FlutterDev team. Drop by to say hi and talk all things open source!

[Google Open Source @GoogleOSS]

🗓️ https://t.co/eqYVze2K4M

[Google Open Source @GoogleOSS]

Where’s the futuristic app we were promised?

Don’t miss Google's keynote at #AllThingsOpen: "Where Is My Futuristic App? Generative UI and the Open Future" with @redbrogdon, @marktechson & @muncus!

[AGI House @agihouse_org]

Framework Highlight: LiteRT ⚡️

The Google Gemma Hardware Hackathon @ AGI House.

Great models need a framework that brings them to your device. Meet LiteRT, Google’s framework for high-performance on-device AI.

At the hackathon, explore how LiteRT can help you turn a Gemma-powered idea into a working local application:

• Run Gemma with LiteRT-LM across edge platforms: https://t.co/QoOhSKMv0Y
• Ready-to-use models for high-performance on-device deployment: https://t.co/tEiKoM8bws
• Explore hardware acceleration on supported devices: https://t.co/4776X3Gnr2

From offline assistants to intelligent edge devices—what will you build?

🎁Best Use of LiteRT: $15,000
Sponsored by Google AI Edge, this award is part of the event’s $45,000 cash prize pool, alongside partner hardware awards.

Join 100 selected builders, the @GoogleDeepMind Gemma team, and hardware partners for a full day of hands-on building. Every project must use a Gemma-based model and run entirely on-device.

October 24, 2026 · 10 AM–5 PM PT
AGI House · Hillsborough, CA

Apply to build: https://t.co/q4l99km6yk
#LiteRT #Gemma #GoogleAIEdge #OnDeviceAI #AGIHouse

[Gophercon Africa @gophers_africa]

🤖 Building agentic apps? Check out Genkit.

@Google open source AI SDK for building full-stack agentic experiences

Test locally with the Developer UI, then deploy anywhere your code runs.

https://t.co/tE2EpS1E4c

#Golang #AI

[Google Open Source @GoogleOSS]

@emckean @thenewstack @linuxfoundation Read how Docsy (used by 2,200+ projects including Kubernetes, OpenTelemetry & gRPC) is evolving for both human developers and AI agents: https://t.co/bIo9dhlz9S

[Google Open Source @GoogleOSS]

"Open source projects work best when they are close to the users."

@emckean spoke with @thenewstack on Docsy joining @linuxfoundation—plus how llms.txt & agent-friendly doc scores reduce maintainer toil.

👇 Link in first reply

[Google Gemma @googlegemma]

Welcome to the new Gemma documentation!

We've built a brand new home for developers to explore, experiment, and build.

What’s new?
🌐 Global model selector for customized code snippets
🤖 Better agent support (turn any page into Markdown!)
📊 Data-driven backend for a consistent experience
⚡️ On-device demos running directly in your browser!

Check it out: https://t.co/yx9K4xGH8M

[Flutter @FlutterDev]

☕ Slicing server bills with fullstack Dart → https://t.co/w2yXSS1bL8

Learn how replacing Node.js Cloud Run services with a single Dart function brought end-to-end type safety and major cost savings across the stack.

[Google Open Source @GoogleOSS]

@darknesspwn @Sophia_IV Yes, but that's a little recursive for a true metaphor 😀

[Google Open Source @GoogleOSS]

Open source is free as in puppies, not free as in beer. 🐶📦

While prior research focuses on upstream maintenance, @sophia_iv studied the effort required to USE and update open source packages at Google.

👇 Link in first reply

[Technik Sam @TechnikSam]

Google just launched https://t.co/6grtTpFNKA — the new central home for Gemma open models 🚀

Features interactive visual architectures, fine-tuning recipes, and instant browser-native demos!

#Gemma #OpenSource #GoogleAI #MachineLearning