FreakyJolly: AI Tutorials, ComfyUI Workflows and Local AI Guides
Siri AI Opens to Every App: iOS 27.3 Developer Preview, Timeline, Limits and How It Works
Fizgig 9.2: Train a MOSS-Image LoRA in 20 Minutes (Install, Settings, VRAM and ComfyUI)
RTX 3060 12GB for AI (2026): What It Runs, Real Speeds, Price and Is It Still Worth It
NVIDIA DGX Spark for Local AI (2026): Price, 64 GB vs 128 GB, Real Speeds and Setup
Gemini in Google Docs Guide (2026): Help Me Create, Match Writing Style, Audio and Docs Live
Arete-27B: Nous Research’s Qwen3.8 Fine-Tune Explained, Benchmarks and How to Run It Locally
The world of AI. Explained.
A clearer view of artificial intelligence.
Hands-on guides for ComfyUI workflows, AI video, image editing and running models locally — tested, with every file path and setting written down.
- Lead story
Siri AI Opens to Every App: iOS 27.3 Developer Preview, Timeline, Limits and How It Works
Apple opened Siri AI to all third-party apps in the iOS 27.3 betas. What you will be able to ask Siri, when it reaches users, EU and language…
The reading list
Latest stories
Fizgig 9.2: Train a MOSS-Image LoRA in 20 Minutes (Install, Settings, VRAM and ComfyUI)
Fizgig 9.2 cuts MOSS-Image LoRA training from 49 to about 20 minutes and batch-3 memory from 21.7 GB to 7.1 GB. What…
Apple Intelligence on Mac Guide (macOS 27): Siri AI, Write With Siri, Supported Macs and How to Turn It On
Every Apple Intelligence feature on the Mac in macOS 27.2: Siri AI and the Siri app, Write with Siri, Visual Intelligence, Image…
RTX 3060 12GB for AI (2026): What It Runs, Real Speeds, Price and Is It Still Worth It
Is the RTX 3060 12GB still worth it for local AI in 2026? Real tokens per second, which LLMs and image models…
NVIDIA DGX Spark for Local AI (2026): Price, 64 GB vs 128 GB, Real Speeds and Setup
Is the NVIDIA DGX Spark worth it for local AI in 2026? Specs, the $6,950 price and new 64 GB model, llama.cpp…
Gemini in Google Docs Guide (2026): Help Me Create, Match Writing Style, Audio and Docs Live
Every Gemini feature in Google Docs in 2026: Help me create, Refine, Match writing style, Match doc format, audio summaries and Docs…
Audio8-TTS-Realistic: Run the 0.6B Voice Cloning Model Locally in 23 Languages (Setup, Scores, Licence)
Edge0’s Audio8-TTS-Realistic is a 0.6B Apache 2.0 text-to-speech model that clones a voice from one clip in 23 languages, including Hindi. Per-language…
Arete-27B: Nous Research’s Qwen3.8 Fine-Tune Explained, Benchmarks and How to Run It Locally
Nous Research’s Arete-27B retrains Qwen3.8-27B to cut state propaganda from 22% to 0.9% (vendor-reported). Full numbers, caveats, GGUF sizes per GPU, Ollama,…
OpenAI Agents SDK Python Guide (2026): Install, Tools, Handoffs, Guardrails and Tracing
Build AI agents with the OpenAI Agents SDK 0.24 in Python: install, first agent, tools, MCP, handoffs, guardrails, sessions and tracing with…
Reasoning Effort Settings: How Thinking Levels Change API Cost (Claude, OpenAI, Gemini)
How Claude effort, OpenAI reasoning.effort and Gemini thinking_level change cost and quality, with Vals AI data, Python code and a decision table.
How every guide is built
Copy the command. Drop the file. Run the graph.
Each ComfyUI guide follows the same shape, so you can skim straight to what you need.
- Step 1InstallNode packs and exact download commands.
- Step 2Files & pathsEvery model folder in one table.
- Step 3SettingsStarting values that work on consumer GPUs.
- Step 4TroubleshootKnown errors and the fix for each.
New models land fast. Get the working setup first.
Follow FreakyJolly in any feed reader to get every new guide the moment it goes live.