August 8, 2026
K3Flight runs the 2.8T-parameter Kimi K3 model on CPU with only ~55GB RAM by streaming weights from storage via cPilot Runtime. A single-file Linux inference server.
Explore a curated directory of open-source AI agent skills for growth hacking, marketing, and revenue operations, covering everything from SEO to lifecycle marketing.
doc7 converts PDFs, Office files, scans, and diagrams into AI-ready Markdown using your own multimodal model, eliminating OCR stacks and per-page fees.
探索竹知了项目：一个零依赖单文件的 Web 模拟，真实录音采样、绳系质点物理，移动端优先，带你重温童年玩具的声音。
Discover how a 2.78-trillion-parameter Kimi K3 model runs on a single CPU with just 8.24 GB of RAM, using portable C99 and clever memory streaming.
Inflect v2 delivers complete 24 kHz English TTS in two compact models (3.96M and 9.36M params) with no external vocoder. Explore architecture, evaluation, and CPU deployment.
KittenTTS delivers high-quality text-to-speech in models as small as 25MB, running entirely on CPU. Learn how to use it, its API, and why it matters for edge AI.
CrispASR is a unified C++ speech engine built on ggml, supporting 53 ASR and 51 TTS models with zero Python dependencies. Run Cohere Transcribe, Parakeet, Voxtral, Qwen3, and more from a single CLI.
Learn how to train a 65M-parameter Vision-Language Model from scratch in just 2 hours for under $3, with full code and dataset details.
Learn how to use the geo-seo-claude skill for Claude Code to audit and optimize websites for AI-powered search engines like ChatGPT, Claude, and Perplexity.
OpenDrop is a Python command-line tool that implements the Apple AirDrop protocol, enabling file sharing with iOS and macOS devices over Wi-Fi.
Learn how Per-Layer Embeddings from Google's Gemma models enable a 28.9M parameter language model to run entirely on an $8 ESP32-S3 chip, generating text at 9.5 tokens per second.