ZeroModels is a collection of pretrained models built entirely in Keras 3. It spans a broad range of tasks, including image classification, object detection, segmentation, monocular depth estimation, feature extraction, vision-language modeling (VLMs), speech recognition, speech-aware language modeling, text encoding and masked language modeling, large language models (LLMs), text-to-text encoder-decoder modeling, multimodal vision-language generation, and more.
From PyPI (recommended)
pip install -U zeromodelsFrom Source
pip install -U git+https://github.com/IMvision12/ZeroModelsDetailed guides are available for all supported tasks, with architecture notes, usage examples, pretrained weights, and real model outputs.
Classification backbones share a single documentation page, while other model families have dedicated pages.
Documentation sources are also available in docs/.
-
Text Encoders (text → embeddings, masked LM, classification)
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights BERT BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding transformersModernBERT Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder transformersELECTRA ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators transformersRoBERTa RoBERTa: A Robustly Optimized BERT Pretraining Approach transformersXLM-RoBERTa Unsupervised Cross-lingual Representation Learning at Scale transformersDeBERTa DeBERTa: Decoding-enhanced BERT with Disentangled Attention transformersDeBERTa-v2 DeBERTa: Decoding-enhanced BERT with Disentangled Attention transformersDeBERTa-v3 DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing transformersT5 Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer transformers -
Text LLMs (text → text)
-
Backbones
-
Object Detection
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights D-FINE D-FINE: Redefine Regression Task of DETRs as Fine-grained Distribution Refinement transformersDETR End-to-End Object Detection with Transformers transformersTable Transformer PubTables-1M: Towards comprehensive table extraction from unstructured documents transformersEfficientDet EfficientDet: Scalable and Efficient Object Detection automlRT-DETR DETRs Beat YOLOs on Real-time Object Detection transformersRT-DETRv2 RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformers transformersRF-DETR RF-DETR: Neural Architecture Search for Real-Time Detection Transformers transformersOWL-ViT Simple Open-Vocabulary Object Detection with Vision Transformers transformersOWLv2 Scaling Open-Vocabulary Object Detection transformersGrounding DINO Marrying DINO with Grounded Pre-Training for Open-Set Object Detection transformers
-
Segmentation
-
Feature Extraction
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights DINO Emerging Properties in Self-Supervised Vision Transformers torch.hubDINOv2 DINOv2: Learning Robust Visual Features without Supervision transformersDINOv3 DINOv3: Self-Supervised Visual Representation Learning at Scale transformers(gated)
-
Depth Estimation
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights Depth Anything V1 Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data transformersDepth Anything V2 Depth Anything V2 transformersTIPSv2-DPT TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment transformers
-
Vision-Language Encoders
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights CLIP Learning Transferable Visual Models From Natural Language Supervision transformersMetaCLIP 2 MetaCLIP 2: A Worldwide Scaling Recipe transformersSigLIP Sigmoid Loss for Language Image Pre-Training transformersSigLIP2 SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features transformersTIPSv2 TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment transformers
-
Multimodal LLMs (image + text → text)
-
Vision-Language Grounding (object detection, OCR, pointing, referring)
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights LocateAnything LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding transformers
-
Speech (speech → text)
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights Whisper Robust Speech Recognition via Large-Scale Weak Supervision transformersSpeech2Text fairseq S2T: Fast Speech-to-Text Modeling with fairseq transformersMoonshine Moonshine: Speech Recognition for Live Transcription and Voice Commands transformersGranite Speech 5 Granite-speech: open-source speech-aware LLMs with strong English ASR capabilities transformers
-
Speech LLMs (audio + text → text)
🏷️ Model Name 📜 Reference Paper 📦 Source of Weights Granite Speech Granite-speech: open-source speech-aware LLMs with strong English ASR capabilities transformersGranite Speech Plus Granite Speech Plus Technical Report transformers
This project leverages timm and transformers for converting pretrained weights from PyTorch to Keras. For licensing details, please refer to the respective repositories.
- 🔖 zeromodels Code: This repository is licensed under the Apache 2.0 License.
- The Keras team for their powerful and user-friendly deep learning framework
- The Transformers library for its robust tools for loading and adapting pretrained models
- The pytorch-image-models (timm) project for pioneering many computer vision model implementations
- All contributors to the original papers and architectures implemented in this library
@misc{gc2025zeromodels,
author = {Gitesh Chawda},
title = {ZeroModels},
year = {2025},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/IMvision12/ZeroModels}}