FreakyJolly
AI Tutorials, ComfyUI Workflows & Local AI Guides
- Website
- freakyjolly.com · all feeds from this site
- Feed URL
https://www.freakyjolly.com/feed/
Read it in your FeedListing reader
Or add to: Feedly Inoreader Raw feed
Latest posts
-
Blender MCP With a Local LLM: Build Animations in VS Code Cline With Qwen3.8-Flash-Next (2026 Setup Guide)
Run Blender MCP with a local LLM: prerequisites, Qwen3.8-Flash-Next and smaller tool-calling models, exact Cline, Continue and OpenCode config JSON, and a worked prompt-to-animation example.
-
Qwen-Image 2.1 Image Edit in ComfyUI (2026): 8 to 16 GB Setup, Multi-Image Edits and Best Edit LoRAs
Qwen-Image 2.1 is not just a text-to-image model. It is a single 7B model that also edits images, and since its September release it has quietly become the best open image editor you can run at home: up to 10 reference images in one edit, painted marks to show exactly what to…
-
LTX-2.5 in ComfyUI on 16 GB (2026): GGUF Setup, Heretic Text Encoder and Best LoRAs
LTX-2.5 from Lightricks is the open video model most people want to run right now. It is a 22B model that makes video and synchronized audio in one pass, it has an 8-step distilled version, and since August Lightricks has released more than a dozen official LoRAs for it, from…
-
Wan 2.2 in ComfyUI on 8 to 16 GB (2026): Fast 4-Step Setup, GGUF Files and Best LoRAs
Run Wan 2.2 image-to-video and text-to-video on 8 to 16 GB GPUs: GGUF files, LightX2V 4-step LoRAs (1022 and 1217), the Slow-Mo Fix, exact ComfyUI folders and the best creative LoRAs.
-
Uncensored Local AI Video Generation in ComfyUI (2026): Wan 2.2, LTX-2.5, MiniMax H3 and Kandinsky 6
The best open video models for creative freedom in ComfyUI, which text encoders hold them back, and the Heretic and abliterated encoders for LTX-2.5, LTX-2.3 and MiniMax H3, with files and folders.
-
Uncensored Local AI Image Generation in ComfyUI (2026): Best Models and Heretic Text Encoders
Why local image models still refuse prompts and how to fix it: Heretic text encoders for Qwen-Image 2.1, FLUX 3 Klein, Z-Image, MOSS-Image and Ideogram 4, with files, nodes and ComfyUI folder paths.
-
Best Uncensored Local LLMs in 2026: Abliterated and Heretic Models by VRAM
The best uncensored local LLMs of November 2026 (abliterated, Heretic and fine-tuned) by VRAM, with Hugging Face repos, GGUF picks and Ollama and llama.cpp commands.
-
Best Local LLMs With Vision and Video Understanding for Consumer GPUs (November 2026)
The best local vision and video LLMs for 8 to 32 GB GPUs: Qwen3.8, MiniCPM-V 4.7, Qwen3.6, Gemma 4, Qwen3-VL, LFM3 and more, with GGUF picks and Ollama and llama.cpp video steps.
-
Ollama Commands Cheat Sheet: Every CLI Command, Modelfile Setting, Environment Variable and API Endpoint
Every Ollama command in one place: CLI, chat slash commands, Hugging Face GGUF, Modelfile, parameters, environment variables, context length and the API, checked against current Ollama.
-
ComfyUI Samplers and Schedulers Explained: Which to Use for Every Model (2026)
Every ComfyUI sampler and scheduler explained, plus steps, CFG, denoise and shift, with the best settings for SD 1.5, SDXL, Flux, Qwen Image, MOSS, Wan and MiniMax H3.
-
ComfyUI Command Line Arguments: Every Launch Flag Explained (2026 Cheat Sheet)
Every ComfyUI launch flag explained: VRAM, Dynamic VRAM, precision, Sage Attention, caching, --listen, folders, previews and custom nodes, with copy-paste presets for 6 to 24 GB GPUs.
-
GGUF Quantization Explained: Q4_K_M vs Q5_K_M vs Q8_0 and Which File to Download for Your VRAM
What every GGUF name means (Q4_K_M, IQ3_XS, Q8_0), how to calculate file size and VRAM including the KV cache, and which quant to pick for 8, 12, 16, 24 and 32 GB GPUs, for LLMs and ComfyUI.
-
ComfyUI Errors and Fixes: 22 Common Errors Explained (Complete 2026 List)
The complete list of common ComfyUI errors with causes and fixes: missing nodes, value not in list, CUDA out of memory, mat1 and mat2, deserializing header, black images, launch flags and more.
-
ComfyUI Model Folder Cheat Sheet: Where Every Model File Goes (2026)
Every ComfyUI model folder explained: which loader reads it, what file goes there, how to tell from a file name, exact paths for popular 2026 models, GGUF, custom node folders and shared model paths.
-
Pruna P-Q-Image-2.1: Qwen Image 2.1 in 6 Steps With ComfyUI LoRAs (v0.2 Guide)
Pruna's P-Q-Image-2.1 v0.2 LoRAs run Qwen Image 2.1 in 6 steps with CFG off, up to 6.3x faster. Which LoRA to pick, full folder paths, ManualSigmas values, editing and 16 GB tips.
-
Krea 2 Turbo in 3 Steps: The 3-Step Distill LoRA for ComfyUI (2.6x Faster)
The Krea2-Turbo-Distill-3step-LoRA makes Krea 2 Turbo about 2.6x faster in ComfyUI while keeping fine texture. Files, folder paths, settings, trade-offs and 16 GB tips.
-
Anima Reference Adapter v2: Consistent Anime Characters From One Image in ComfyUI
Anima Reference Adapter v2 keeps an anime character consistent from one reference image, with no LoRA training. Install steps, model paths, best settings, two-character prompts and speed.
-
ComfyArt: Free Local AI Design Canvas for ComfyUI (Install, Models and Agent Guide)
ComfyArt is a free, MIT-licensed AI design canvas that runs inside ComfyUI. Install steps, every model with its full folder path, the design agent, MCP and Comfy Cloud setup.
-
Liquid AI LFM3: How to Run the 2B Text, Vision and Audio Models Locally (Ollama, llama.cpp, LM Studio)
LFM3 is Liquid AI's new family of six open-weight 2B models. See the benchmarks, file sizes and exact Ollama, llama.cpp and LM Studio commands to run LFM3-1.7B, LFM3-VL-2B and the audio models locally.
-
StableGen 0.4 for Blender: AI Texturing, Auto-Rigging and Mixamo Animation With ComfyUI (Install Guide)
StableGen 0.4.0 textures, generates and now auto-rigs 3D characters in Blender using ComfyUI. Installer packages, custom nodes, exact model paths, VRAM needs and licences explained.