Resource directory
Find what you need for your workflow.
Guides: techniques, setup, and troubleshootingSearch Sweet Tea Pipes, external workflows, models, node packs, and guides in one evidence-linked catalog.
Start with a workflow instead ↗Search results
All resources
20 shown of 41,789
Merger_alice_antsime
Another merger
ZONOS2 GGUF
GGUF builds of Zyphra/ZONOS2 — an 8B-parameter (SonicMoE, 900M active) real-time text-to-speech model with voice cloning — for mistral.rs.
Yulan Style
Works best at 0.75 weight.
Faline
oc by Tip
Unlimited OCR Mixed NVFP4
A Mac-native (Apple Silicon / MLX) mixed-precision build of baidu/Unlimited-OCR (built on DeepSeek-OCR), packaged for the Krill runtime. It parses documents and images to grounded text with no Python and no...
XL Sato Shin/佐藤心/佐藤心 Idolmaster/偶像大师
Base Model: Animagine XL V3.1
ComfyUI-MiniMaxH3-PerRowMasking
Temporary MiniMax H3 per-row generation masking shim for ComfyUI
Merger_minliin_utah
Another merger from my TIs
Hio Makoto Style
Works best at 0.75 weight.
[PonyV6 XL LoCon] Reika Shiki | Onmyo Knight Towako
This is a requested character.
Idolomantises artist style
https://www.patreon.com/idolomantises https://x.com/Idolomantises
DeepFilterNet3 Streaming CoreML
Same DeepFilterNet3 Neural Engine model as aufklarer/DeepFilterNet3-CoreML, re-exported with the three SqueezedGRU hidden states as explicit inputs and outputs. Carrying those states across blocks gives glitch-free...
Tacfigure Style (Anima & Illustrious)
Works best at 0.95 weight.
Pintail (Kanatale / Swordgirls)
Thanks Zaciel for training data
Artist:ame usari 飴乃うさり(NoobXL)
偏少女的一款画风(?)
XL Kohinata Miho/小日向美穗/小日向美穂 Idolmaster/偶像大师
Base Model: Animagine XL V3.1
Merger_marie_omaly
Another merger from my TI's
Ligod Style
Works best at 0.75 weight.
MMEvol Qwen2 7B
Here are the pretrained weights and instruction tuning weights | Model | Pretrained Projector | Base LLM | PT Data | IT Data | Download | | MMEvol-Qwen2-7B | mmprojector | Qwen2-7B | LLaVA-Pretrain | MMEvol | ckpt |
Qwen3.5 9B GGUF
This is a GGUF-quantized version of Qwen3.5-9B produced with ByteShape's ShapeLearn, which learns the optimal datatype per tensor to maintain high quality even at very low bitlengths.
