Huawei releases MACaQ, its internal quantizer for local inference (github.com)

Screenshot of github.com: Huawei releases MACaQ, its internal quantizer for local inference

First screen, captured 2 days ago. Open the video.

Full page

Captured 2 days ago. Click for the whole screenshot.

Looks similar

Screenshot of github.com: Octomind – open-source AI harness that saves your tokens, static rust

Octomind – open-source AI harness that saves your tokens, static rust (github.com)

41 points by donk8r 10 days ago | 32 comments

GitHub repository for Octomind, an open-source Rust AI coding agent and runtime that works with any model, is MCP-native, and runs in terminal, CI, or as a daemon.

ai-coding-agent rust mcp llm

Screenshot of github.com: Show HN: Recalc – Spreadsheet where cells can be Claude Code or Codex terminals

Show HN: Recalc – Spreadsheet where cells can be Claude Code or Codex terminals (github.com)

20 points by sshah03 5 days ago | 3 comments

GitHub repo for recalc, an open-source desktop spreadsheet where cells can hold values, formulas, terminal commands, or Claude Code/Codex agent sessions.

claude-code codex ai-agents spreadsheet

Screenshot of github.com: Show HN: Cerulion – ROS 2-compatible robotics runtime with multi-process replay

Show HN: Cerulion – ROS 2-compatible robotics runtime with multi-process replay (github.com)

8 points by lakshays 1 day ago | 2 comments

GitHub repository for Cerulion, an AGPL-licensed Rust robotics runtime compatible with ROS 2 offering zero-copy shared-memory IPC and deterministic multi-process replay.

ros2 robotics rust iceoryx2

Screenshot of github.com: 6-11X speedup running Qwen3.6 on my RTX 4090, full repo.

6-11X speedup running Qwen3.6 on my RTX 4090, full repo.(github.com)

10 points by AI4Science 6 hours ago | 0 comments

GitHub repository for Lightreason, a Python/CUDA project claiming LLM token-generation speedups for Qwen models on consumer GPUs, with benchmarks and a demo chat.

llm-inference speculative-decoding qwen rtx-4090

Screenshot of github.com: Show HN: Ramasite, a Self-Contained Laravel/Filament/SQLite CMS

Show HN: Ramasite, a Self-Contained Laravel/Filament/SQLite CMS (github.com)

3 points by DaleFranks 1 day ago | 0 comments

GitHub repository for RamaSite, a self-contained MIT-licensed CMS built with Laravel, Filament and SQLite, largely written with Claude Code.

cms laravel filament sqlite

Screenshot of github.com: Show HN: Yogurt – Free, private, open-source alternative to Granola

Show HN: Yogurt – Free, private, open-source alternative to Granola (github.com)

267 points by brownrout 2 days ago | 99 comments

GitHub repository for Yogurt, a free MIT-licensed Electron desktop app that transcribes and summarizes meetings using on-device AI models.

meeting-notes local-ai whisper llama-cpp

Full page

Looks similar

Octomind – open-source AI harness that saves your tokens, static rust (github.com)

Show HN: Recalc – Spreadsheet where cells can be Claude Code or Codex terminals (github.com)

Show HN: Cerulion – ROS 2-compatible robotics runtime with multi-process replay (github.com)

6-11X speedup running Qwen3.6 on my RTX 4090, full repo.(github.com)

Show HN: Ramasite, a Self-Contained Laravel/Filament/SQLite CMS (github.com)

Show HN: Yogurt – Free, private, open-source alternative to Granola (github.com)

Related stories

Kauldron: Modular, scalable library to train ML models (github.com)

Phi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoning (techcommunity.microsoft.com)

Train Your Own LLM from Scratch (github.com)

Fine tune a 70B language model at home (answer.ai)

GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance (github.com)

AWS Bedrock to require sharing data with Anthropic for Mythos and future models