MossyModels is the largest community library for AI aimbot models and YOLO detection models. Browse, download, and share detection models built for aim assist, target tracking, and game automation across dozens of popular first-person shooter games. Our growing collection features hundreds of models uploaded by the community, covering everything from Fortnite to Valorant to Call of Duty.
Find AI detection models organized by game. Each game page lists community-uploaded models you can download for free after creating an account. Models are available in ONNX and PyTorch (.pt) formats compatible with popular object detection frameworks. Whether you need a model for Fortnite, Warzone, Apex Legends, or any other supported title, MossyModels has you covered with a wide selection of community-tested options.
AI aimbot models are machine learning models trained to detect players, enemies, and other targets in video games using real-time computer vision. These models are typically based on the YOLO (You Only Look Once) architecture, a state-of-the-art object detection framework developed by Ultralytics that excels at fast, accurate detection. YOLO models can identify and locate game objects like player outlines, health bars, and character models within individual video frames at high speed, making them ideal for real-time gaming applications.
The detection pipeline works by capturing game frames, feeding them through a trained neural network, and outputting bounding box coordinates around detected targets. Different models are trained on different games since each game has unique character designs, color palettes, map styles, and UI elements that the model needs to learn to recognize accurately.
ONNX (Open Neural Network Exchange) is an open standard format for representing machine learning models. ONNX models offer excellent cross-platform compatibility and can run on various hardware accelerators including NVIDIA GPUs via TensorRT, AMD GPUs, and even CPU-only setups. Most detection frameworks and aim assist tools support ONNX as their primary input format, making it the most versatile choice for end users who want broad compatibility.
PyTorch (.pt) models are the native format used during training with the YOLO framework. These files contain the full model architecture and trained weights. They can be used directly with PyTorch-based inference scripts or converted to ONNX for deployment in production applications. PyTorch models are ideal for users who want to fine-tune or retrain existing models on their own custom datasets.
Uploads must be a PyTorch checkpoint (.pt), ONNX (.onnx), or a zip whose members are only .pt and .onnx files.
Getting started on MossyModels is simple and free. Create an account using the sign up page, then browse models by game. Click on any game to see all available detection models, complete with download counts, average ratings, and version numbers so you can quickly identify the best options. When you find a model you want to try, just hit download and the file will be saved to your computer ready for use.
Every model page shows detailed information including the uploader, upload date, file format, user ratings, and community comments. You can rate models after downloading them and leave feedback to help other users make informed decisions about which models perform best for their specific needs and hardware setups.
Trained a new YOLO detection model? Upload it to MossyModels and share your work with thousands of active users. You can upload ONNX, PyTorch (.pt), or a zip archive that contains only those file types. Add a detailed description, select the target game, choose relevant tags, and optionally include a preview image showing your model's detection results. Models with clear descriptions and preview images get significantly more downloads.
MossyModels is built around its community. Rate models on a five-star scale after downloading and testing them. Write comments to share your experience, report issues, or ask questions about configuration. Top-rated models are featured prominently on game pages so the best community contributions get the visibility and recognition they deserve. Active uploaders and helpful community members build reputation over time.
Yes. Creating an account, browsing models, downloading files, and uploading your own models is completely free. There are no premium tiers, download limits, or hidden fees.
MossyModels supports dozens of popular games including Fortnite, Valorant, Warzone, Apex Legends, Counter-Strike 2, Overwatch 2, Rainbow Six Siege, Rust, The Finals, PUBG, XDefiant, Delta Force, Deadlock, Marvel Rivals, and many more. New games are added regularly based on community requests and uploads.
Look at the download count, average rating, and read community comments. Higher-rated models with more downloads are generally more reliable. Check the model description for details about training data size, resolution, and recommended settings. Models trained on larger datasets (measured in thousands of images, noted as "k" in model names) tend to perform more accurately.
Yes. If the game you want is not listed, you can submit a game request through the site and an administrator will review and add it to the library so models can be uploaded for it.