Faster Gemma 4 on MLX with multi-token prediction
Gemma 4 is now significantly faster in Ollama 0.31 on Apple Silicon via multi-token prediction (MTP), powered by MLX. Performance is now up to 90% faster…
Benchmarking Self-Hosted Gemma 2 9B vs. Frontier APIs: The FP8 Quantization Prefill Tax and VRAM Realities on an NVIDIA L4 [P]
When evaluating migrating production LLM workloads off commercial cloud APIs, the conversation usually gets oversimplified into a trade-off between quality and infrastructure cost. To look past…
Orthrus (diffusion head) trained Qwen 3.5/3.6 and Gemma 4 models are dropping soon
"Hi all, we are finalized with our testing and are preparing the release pipeline. We will be releasing support for the Qwen3.5, Qwen3.6, and Gemma4 very…
Gemma 4 12b needs glasses
Having a lot of fun using Gemma 4 as an assistant, but is growing frustrated with the poor default image resolution setting for image vision. Tasks…
RT Google Gemma: Gemma 4 just hit 200M downloads in only 2.5 months! For context, total downloads across the entire Gemma family of models were at 100…
RT Google GemmaGemma 4 just hit 200M downloads in only 2.5 months!For context, total downloads across the entire Gemma family of models were at 100M when…
RT Google Gemma: Gemma 4 just hit 200M downloads in only 2.5 months! For context, total downloads across the entire Gemma family of models were at 100…
RT Google GemmaGemma 4 just hit 200M downloads in only 2.5 months!For context, total downloads across the entire Gemma family of models were at 100M when…
Worse quality with MTP – Qwen 3.6, Gemma 4
Hi. I am self-hosting Qwen 3.6 27B Q8_K_XL with Llama.cpp on 4x5070ti. (All 4 cards are on single x16 slot bifurcated to 4x4 with risers). I've…
Gemma 4 26BA4B Surprisingly Usable at IQ3_S – Are small quants really this usable?
I've been experimenting with using lower quants of Gemma 4 26B on my M3 16gb MacBook Air. The Quant runs at a solid 25 tokens per…
Build Cross-Language Multi-Agent Team with Google’s Agent Development Kit and A2A
How a Python agent and a Go agent collaborate on contract compliance using the Agent2Agent protocolY...
Measuring What Matters with Jules
AI coding agents are rapidly shifting from reactive assistants that complete tasks when prompted to ...
How A2A is Building a World of Collaborative Agents
Celebrating the first anniversary of the Agent-to-Agent (A2A) protocol, this blog post highlights how the framework enables autonomous AI agents to securely collaborate and hand off…
A2UI + MCP Apps: Combining the best of declarative and custom agentic UIs
This post introduces three architectural patterns designed to integrate Model Context Protocol (MCP) Apps and Agent-to-User Interface (A2UI) to solve the tradeoff between highly custom iframe…
Announcing the Agentic Resource Discovery specification
An open specification for finding and verifying tools, skills, and agents across the web.Agents are ...
RT Leandro von Werra: We launched an agent collaboration with a simple task: make Gemma 4 faster. Over 100 agents from all over the world joined, exch…
RT Leandro von WerraWe launched an agent collaboration with a simple task: make Gemma 4 faster.Over 100 agents from all over the world joined, exchanged 1000+…
Unlocking the Power of the TPU Stack: Introducing our new Developer Hub
Google has officially launched the TPU Developer Hub, a centralized educational resource designed to help model builders and developers maximize the performance of Google Cloud TPUs.…
Introducing Gemma 4 models on Amazon Bedrock
Today, we are announcing the availability of the Gemma 4 family on Amazon Bedrock. Built by Google DeepMind and released under the Apache 2.0 license, Gemma…
Awesome to see this innovation in text diffusion. DiffusionGemma is lightning fast, 4x faster than other Gemma 4 models! Congrats to @bodonoghue85 and…
Awesome to see this innovation in text diffusion. DiffusionGemma is lightning fast, 4x faster than other Gemma 4 models! Congrats to @bodonoghue85 and the team who…
DiffusionGemma: The Developer Guide
DiffusionGemma is an experimental text-generation model built on the Gemma 4 architecture that uses diffusion-based parallel generation instead of token-by-token autoregression, enabling much faster inference, bidirectional…
Introducing the Google Colab CLI
Google has announced the Google Colab Command-Line Interface (CLI), a new tool that allows developers and AI agents to connect local terminals to remote Colab runtimes…
Celebrating the milestone of a massive 150+ million downloads of Gemma 4 with the release of the new Gemma 4 12B model! It's incredibly powerful for s…
Celebrating the milestone of a massive 150+ million downloads of Gemma 4 with the release of the new Gemma 4 12B model! It's incredibly powerful for…
Bringing Gemma 4 12B to your Laptop: Unlocking Local, Agentic Workflows with Google AI Edge
Google DeepMind’s Gemma 4 12B model brings agentic, multimodal AI capabilities to everyday laptops with 16GB of RAM, enabling local data processing and visual insight generation.…
Gemma 4 12B: The Developer Guide
The newly released Gemma 4 12B is a dense, multimodal model designed for high-performance local AI execution on consumer devices. By introducing a novel, encoder-free architecture,…
RT Nathan Lambert: Gemma 4 adoption numbers outpacing Qwen 3.5/3.6 for the same sized models is a big shift in the international balance of influence …
RT Nathan LambertGemma 4 adoption numbers outpacing Qwen 3.5/3.6 for the same sized models is a big shift in the international balance of influence via open…
Latest open artifacts (#21): Open model bonanza! Gemma 4, DeepSeek V4, Kimi K2.6, MiMo 2.5, GLM-5.1 & others. On CAISI's V4 assessment.
An eventful month with one flagship release after another