GoKawiil - Tech News

1.

Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac (news.ycombinator.com)

2026-07-29 | tags: gemma 4 26b, apple silicon, swift metal

2.

Running a 28.9M parameter LLM on an $8 microcontroller (news.ycombinator.com)

2026-07-25 | tags: esp32-s3, gemma models, per-layer embeddings

3.

Show HN: Cactus Hybrid: We taught Gemma 4 to know when it's wrong (news.ycombinator.com)

2026-07-22 | tags: gemma 4, cactus hybrid, huging face

4.

Gboard could soon be able to understand sign language with your phone’s camera (androidauthority.com)

2026-07-17 | tags: gboard, google deep mind, sign language

5.

Running Gemma 4 26B at 5 tokens/sec on a 13-year-old Xeon with no GPU (news.ycombinator.com)

2026-07-15 | tags: build, gemma, hardware

6.

Running Gemma 4 26B at 5 tokens/SEC on a 13-year-old Xeon with no GPU (news.ycombinator.com)

2026-07-15 | tags: build, gemma, hardware

7.

Google’s latest Pixel AI push keeps your data on your phone (androidauthority.com)

2026-07-14 | tags: apps, device, gemma

8.

HackerRank open sourced its ATS. My resume scored 90/100. Oh wait 74. No – 88 (news.ycombinator.com)

2026-06-29 | by Dan Kinsky | get HackerRank ATS Tool → | tags: hackerrank, hiring-agent, open-source

9.

Ornith-1.0: Self-scaffolding LLMs for agentic coding (news.ycombinator.com)

2026-06-28 | get AI Coding Assistant Tool → | tags: ornith-1.0, gemma, quwen

10.

Liquid AI's smallest model yet LFM2.5-230M beats models 4X its size at data extraction, can run 'anywhere' (venturebeat.com)

2026-06-25 | get Portable AI Model Device → | tags: liquid ai, lfm2.5-230m, qwen3.5

11.

I wanted Google’s secret AI dictation app to replace Wispr Flow, but it couldn’t (androidauthority.com)

2026-06-25 | get Google Pixel Buds Pro → | tags: google, wispr flow, gboard

12.

CrankGPT is a private AI chatbot that’s powered entirely by your hand (androidauthority.com)

2026-06-19 | get AI Handheld Chatbot Device → | tags: crankgpt, raspberry pi 5, squeez labs

13.

Running local models is good now (news.ycombinator.com)

2026-06-16 | get NVIDIA Jetson Nano Developer Kit → | tags: llama.cpp, gpt-oss, lm studio

14.

A satellite just learned to find things on its own — here’s what that means (techcrunch.com)

2026-06-15 | by Tim Fernholz | get Satellite Autonomous Navigation Kit → | tags: satellite, earth observation, yam-9

15.

How to setup a local coding agent on macOS (news.ycombinator.com)

2026-06-12 | by Kyle Howells | get MacBook Pro Developer Kit → | tags: gemma 4, llama.cpp, openai

16.

How to Setup a Local Coding Agent on macOS (news.ycombinator.com)

2026-06-12 | by Kyle Howells | get Mac Developer Command Line Tools → | tags: gemma, gguf, llama

17.

Google's DiffusionGemma generates 256 tokens in parallel and self-corrects as it goes (venturebeat.com)

2026-06-11 | get AI Text Generation Tool → | tags: block, diffusiongemma, generation

18.

DiffusionGemma is Google’s fastest AI yet, but it comes with a big trade-off (androidauthority.com)

2026-06-11 | get NVIDIA GeForce RTX 4090 → | tags: diffusiongemma, google, model

19.

Google DeepMind releases DiffusionGemma, a model that runs local AI 4x faster (arstechnica.com)

2026-06-10 | get NVIDIA RTX 4090 GPU → | tags: autoregressive, diffusiongemma, google

20.

DiffusionGemma: 4x Faster Text Generation (news.ycombinator.com)

2026-06-10 | by Brendan O'Donoghue | get AI Text Generation Software → | tags: diffusion, diffusiongemma, entire

21.

The latest Gemma 4 models use a training trick to slash their on-device memory footprint (androidauthority.com)

2026-06-05 | get TensorFlow Lite Model Kit → | tags: gemma, gemma gemma, memory

22.

Google’s latest on-device AI model is custom-made for your laptop (androidauthority.com)

2026-06-04 | get Laptop AI Assistant Software → | tags: google, gemma 4 12b, moe model

23.

Google AI Edge Gallery launches on macOS, letting Mac users run Gemini models locally (9to5mac.com)

2026-06-04 | by Marcus Mendes | get Google AI Edge Developer Kit → | tags: google ai edge gallery, gemini models, google ai edge eloquent

24.

Google Launches 'Gemma 4 12B' AI Model That Can Run On Your Laptop (slashdot.org)

2026-06-03 | get AI Model Running Laptop → | tags: google, gemma 4 12b, gemini program

25.

Google's new Gemma 4 12B model is designed to run on any laptop with 16GB of RAM (arstechnica.com)

2026-06-03 | get Laptop RAM Upgrade Kit → | tags: google, gemma 4, ai models

26.

Google's new Gemma 4 open AI model is sized for your laptop (arstechnica.com)

2026-06-03 | get Google Tensor Laptop Chip → | tags: google, gemma 4, ai models

27.

Google's new open source Gemma 4 12B analyzes audio, video — and runs entirely locally on a typical 16GB enterprise laptop (venturebeat.com)

2026-06-03 | get Google Gemma 4 12B AI → | tags: google, gemma 4 12b, hugging face

28.

Gemma 4 12B: A unified, encoder-free multimodal model (news.ycombinator.com)

2026-06-03 | by Olivier Lacombe | tags: gemma 4 12b, moe, ai security

29.

A 10 year old Xeon is all you need (news.ycombinator.com)

2026-06-01 | get Intel Xeon E5 Processor → | tags: xeon e5-2620, gemma 4, ollama

30.

A 10 year old Xeon is all you need (for 26B-A4B MTP Drafters without GPU) (news.ycombinator.com)

2026-06-01 | get Intel Xeon E5-2670 → | tags: xeon e5-2620 v4, gemma 4, ollama