Running local AI on the Raspberry Pi 5 taught me why cloud models are still winning
…Llama, Gemma, and Deepseek. I opted to install the smallest version of each model, as I only had 32GB of space on the SD Card. Using LLMs on the SBC was responsive…
Search
Related topics:
Google Gemma
All sources
xda-developers.com
66
developer.nvidia.com
15
blogs.nvidia.com
13
android-developers.googleblog.com
5
androidauthority.com
4
wccftech.com
3
arstechnica.com
3
huggingface.co
3
tomshardware.com
2
9to5mac.com
2
amd.com
2
androidpolice.com
2
Tracked topic
Google Gemma
Gemma is a family of open-weight language models released by Google for text generation and related NLP tasks.
No GPU left behind: Unlocking Efficiency with Co-located vLLM in TRL
…File "/usr/local/lib/python3.11/dist-packages/transformers/models/gemma3/modeling_gemma3.py", line 880, in forward [rank0]: logits = self.lm_head(hidden_states[:, slice_indices, :]) [rank0]: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ [rank0]: File "/usr/local…
I ran local LLMs on Intel's cheapest iGPU, and the results were surprisingly decent
…It ain’t perfect, but it’s a decent secondary LLM server I’ve got a Gemma4-26B-A4B instance that runs on my GTX 1080 24/7, and I use it…
Google's $100 AI Ultra is still overpriced, and here's what I use instead
…Qwen See at Github Replacing the AI is only the first step I found a replacement for everything I use Adding a local LLM is only part of the equation, because Google…
Top stories
I tested 3 local LLMs on my RTX 4070 Ti for real work — only one earned a permanent spot
1d agoI ran a powerful local AI model on my laptop, and it didn’t feel like a compromise
5d agoGoogle's new Gemma 4 12B model is designed to run on any laptop with 16GB of RAM
1w agoI ran Gemma 4 and Qwen 3.5 for the same local tasks, and one pulled miles ahead
1w agoI replaced Cursor and Antigravity with a completely local VS Code setup, and I missed less than I expected
…I already use one workstation for my Proxmox experiments, while the other is my main gaming/video-editing/coding machine. Then there’s the privacy advantage of hooking local LLMs up to…
Google Quietly Launches Free Offline Dictation App for iOS
…performing speech recognition tasks locally. This means users can record and transcribe audio in places with poor or no network access. The app uses Google's Gemma-based speech models to convert…
Local LLMs are actually good now, and I wasted months not realizing it
…Gemma 4 isn't the smartest local LLM I've run, but it's the one I reach for most Google's newest Gemma 4 models are both powerful and useful. The…
Discussions and forums
Using a Gemma4 Safetensor Already Downloaded Locally
Hi everyone. I need some help or advice. I’m learning how to use N8N, so I downloaded Docker and installed N8N locally. I also wanted to install Gemma4, which I use in ComfyUI to help with image generation prompts. Is it…
Qwen 3.6 27B vs Gemma 4 31B - making Packman game!
Gemma just crushed Qwen in a local LLM gamedev contest! Device: MacBook Pro M5 Max, 64GB RAM Qwen 3.6 27B: 32 tokens/sec · 18m 04s · 33,946 tokens. Gemma 4 31B: 27 tokens/sec · 3m 51s · 6,209 tokens. So what is more impo…
Show HN: Hitoku Draft – Context aware local assistant
Hi guys.I have been working on Hitoku Draft, an open-source, voice-first AI assistant that runs entirely locally. I posted about it already, and now it has also transcription with voice editing. Looking for feedback, as …
15
1
Ask HN: Any Local LLM can I run without GPU for Local Agentic workflow AI?
Claude Code like agentic workflow ai too costly for me.Any LLM can I run with VSCode at the below setup? 16ram Intel core i7 h processor 13gen 512gb NVMe SSD I want to run the ai as local agentic workflow with Vscode.I w…
5
2
Multi-Token Prediction (MTP) for LLaMA.cpp - Gemma 4 speedup by 40%
Implemented Multi-Token Prediction for LLaMA.cpp. Quantized Gemma 4 assistant models into GGUF format. Ran tests on a MacBook Pro M5Max. Gemma 26B with MTP drafts tokens 40% faster. Prompt: Write a Python program to find…
Updating Classifier Evasion for Vision Language Models | NVIDIA Technical Blog
…A widely-used open source example is PaliGemma 2 . As shown in Figure 1, PaliGemma 2 uses SigLIP to encode and project the image into token space compatible with Gemma 2, then…
New Adobe Premiere Color Grading Mode Accelerated on NVIDIA GPUs
…by 15%. ✨ Google’s Gemma 4 family of omni-capable models are built for local AI across a wide range of devices. Google and NVIDIA have optimized Gemma 4 for NVIDIA GPUs…
LM Studio now lets you use your iPhone to talk to local models on your Mac - 9to5Mac
LM Studio now lets you use your iPhone to talk to local models on your Mac Marcus Mendes | Jun 4 2026 - 10:21 am PT | Jun 4 2026 - 10:21 am PT…