13 stories tagged with #gemma, in publish-time order across the WeSearch catalog. Tag pages update as new stories ingest.
⌘ RSS feed for this tag → or search "Gemma"
gemma-4-31B-it-DFlash has been released
I guess we'll have to wait until this PR is merged before we can test it.…
GPU server for hosting Gemma 4 possibilities
I stumbled on a Gemma 4 chat template bug for tools and fixed it
great work, Gemma
I've created a LoRA for Gemma 3 270M making it probably the smallest thinking model?
Here is an example of the output: ``` ==================== THINKING ==================== Here is the thinking process: This is a large community with a wide range of interests User…
Cartoonist Gemma Correll's moving and funny book about her lifelong mental illness: Anxietyland
Gemma Correll's Anxietyland covers panic attacks, agoraphobia, depression, and a hospitalization program — and it's also funny.…
AMD Radeon RX 6900 XT - ROCm vs Vulkan - Gemma 4 and Qwen 3.5 speed benchmarks
Did some quick tests after building llama.cpp with ROCm 6.4.2 and latest Vulkan for my 6900 XT gemma4 E2B Q4_K ubatch ROCm pp512 Vulkan pp512 ROCm tg128 Vulkan tg128 32 1536.60 142…
How to Use Transformers.js in a Chrome Extension
We’re on a journey to advance and democratize artificial intelligence through open source and open science.…
How to run a local coding agent with Gemma 4 and Pi
Set up Gemma 4 running in LM Studio, connected to Pi as the terminal agent…
Most efficient way of running Gemma 4 E4B with multimodal capabilities on a laptop?
The gemma 4 E4B and E2B models have built-in multimodal capabilities. However, as far as I am aware, llama.cpp does not have proper support for vision and audio inputs (specially a…
Offline Agentic Coding
Offline Agentic Coding: Ollama and Claude code…
How to run a local coding agent with Gemma 4 and Pi | Patrick Loeber
Tutorial from the Google guy, I use very similar setup (llama.cpp instead of lmstudio)…
Speculative decoding with Gemma-4-31B + Gemma-4-E2B enables 120 - 200 tok/s output speed for specific tasks
So for my project I was using up until now either Gemini 3 / 2.5 Flash or Flash-lite. All my use cases are not agentic, simply LLM workflows for atomic tasks like extracting refere…