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 41,716
Output ControlNet Pth v1
These are controlnet weights trained on runwayml/stable-diffusion-v1-5 with new type of conditioning.
MeshAnything
MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers
Pipevision Yolo26m Seg
Model: yolo26m-seg-pipevision-fp16 Format: ONNX FP16 (opset 17) Input: images [1, 3, 640, 640] — NCHW, float16, RGB normalized [0, 1], letterboxed (pad 114) Outputs: output0 [1, 100, 38] float32 — end-to-end...
Action Detection
This model performs human action classification on videos using a CNN-GRU architecture built on top of MobileNetV2 (1.0, 224) features and trained on the UCF101 dataset. It is well-suited for recognizing actions from...
Unlearn Cl Ucf101 Videomae Large Salun 8 42
This model is a fine-tuned version of MCG-NJU/videomae-large on the ucf101 dataset. It achieves the following results on the evaluation set: Loss: 0.3518 Accuracy: 0.9169 Test Accuracy: 0.9169 Df Accuracy: 0.9450...
Siglip2 Base Patch16 512 Guj 32768
This model is a 45.62% smaller version of google/siglip2-base-patch16-512 optimized for Gujarati language via vocabulary size reduction using the trimming method. This trimmed model should perform similarly to the...
Gigaam Multilingual Large Ctc Onnx
GigaAM Multilingual Large CTC model converted to ONNX format for onnx-asr.
ControlNetCannyReducedImproved
These are controlnet weights trained on black-forest-labs/FLUX.1-dev with new type of conditioning. You can find some example images below.
Robust U1 SFT
Robust-U1 is a unified MLLM that self-recovers corrupted visual content and reasons over it, enabling robust visual understanding under real-world image degradations.
Whisper Tiny Cv11
A plain openai/whisper-tiny (unmodified architecture) fine-tuned on the 11 languages of the CC0 Whispered corpus. This is the matched scientific control for burakaydinofficial/whisper-tiny-mla-cv11 — trained...
ControlNet Streetview Seg Bins Safety
These are controlnet weights trained on runwayml/stable-diffusion-v1-5 with new type of conditioning. You can find some example images below.
Unlearn Ucf101 Videomae Base Scrub 4 42
This model is a fine-tuned version of MCG-NJU/videomae-base on the ucf101 dataset. It achieves the following results on the evaluation set: Loss: 0.7989 Accuracy: 0.8310 Test Accuracy: 0.8310 Df Accuracy: 0.9915...
Omnilingual ASR CTC 1B ONNX
Meta omniASRCTC1Bv2 exported for server-side ONNX Runtime (CUDA EP with CPU fallback). Exported by models/omnilingual-asr/export/convertonnx.py in soniqo/speech-models.
Lcnn
> Improved version available: caa-speech-detection-asvspoof2019/lcnn-v7-cqt — lcnnv7 (CQT + label-smoothing + cosine + grad-clip) achieves 3.26% eval EER (vs 8.43% here) and 0.4930 tandem min t-DCF.
Qwen3 VL 2B Instruct Image To Mermaid
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
Huihui Gemma 4 E2B It Abliterated
This is an uncensored version of google/gemma-4-E2B-it created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from...
Oxford 17 Flowers Cnn
ResNet18 fine-tuned on the Oxford 17 Flowers dataset. Classifies RGB images into 17 flower categories.
Autocatalogai Clip Multitask v2
AutoCatalogAI V2 is a multi-task computer vision model for automatically extracting fashion product attributes and generating structured catalog metadata from a single product image.
Siglip2 So400m Patch16 512 Vision
> Note: This is a stripped-down copy of google/siglip2-so400m-patch16-512 containing only the vision tower. It encodes images / video frames into the shared embedding space and is meant to be loaded on its own at...
Korean Kws2
This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set: Loss: 0.3841 Accuracy: 1.0
