Inbibo has released Adonis 2.1, the latest version of its digital anatomy framework for Autodesk Maya and SideFX Houdini. This update introduces new product segmentation, adds machine learning tools for faster deformation feedback, expands the anatomy transfer workflow, and introduces new deformers for cleaner simulation output. Apparently, simulating muscle, fat, skin, and fascia still takes too long.
Starting with version 2.1, Adonis has been split into two products. AdonisFX remains the core digital anatomy toolset for muscle, fat, and skin simulation in Maya and Houdini. AdonisML is a new high-end version that includes everything in AdonisFX plus two machine learning systems: ML Deformer and SmartTissue.
The overall idea is pretty simple, at least by the insane standards of character effects. This means that animators can express the final character faster and more interactively while working, while maintaining a high-quality anatomical setup.
ML deformers for interactive character feedback
A new ML deformer is trained on AdonisFX characters to create pose-based machine learning approximations of the deformations produced by heavier tissue setups. Inbibo describes the system as a custom neural network training pipeline.
SmartTissue restores secondary movement
A limitation of pose-based ML deformations is that dynamic information is usually lost. Although the trained result may reproduce the shape well, it does not automatically preserve the wobbles, lulls, and overshoots caused by fascia and fat simulations. It’s often this lack of motion that makes creatures and characters feel less like rigs and more like living, slightly expensive problems.
SmartTissue is Inbibo’s answer to that limitation. The system uses a neural network architecture to predict how the underlying muscles are active for the current pose and uses those predictions to drive faster dynamics layers. The goal is to restore secondary movements such that a complete anatomical simulation is required.
According to Inbibo, SmartTissue can also be cached, allowing animators to generate playblasts using dynamics that already exist. The company says that cached results are often enough to move on to lighting and client reviews within minutes for shots that include multiple simulated creatures. The pipeline enjoys grim optimism, so while this won’t eliminate the need for final validation, it should make iterating less difficult.
Anatomical implantation with mismatched topology
AdonisFX 2.1 also provides a topology-independent anatomical structure transfer workflow. The new AdnTransfer system is designed to transfer anatomical geometry from one character or creature to another by mapping the anatomical correspondence between source and target. The important part is that the source and target don’t have to share topology or the same ratios.
This is useful because many vertebrates share roughly equivalent anatomy, even though mesh, scale, and proportions vary widely. Such a transfer workflow can reduce the amount of manual reconstruction required when moving muscle and tissue setups between related characters or creatures.
Two new deformers are at the heart of your workflow. AdnRigidWrap attaches a mesh to one or more target meshes, locking each point to its closest location on the target while preserving the original offset. AdnSoftWrap handles soft regions between these attachments by blending nearby influences within a defined radius, creating a continuous soft deformation across the mesh. A new landmark tool provides a visual interface for placing and configuring matched pairs directly on the mesh.
AdnMush smooths without losing detail
Another major addition is AdnMush, a new deformer to reduce skinning and post-simulation artifacts while preserving anatomical surface details. Inbibo specifically mentions details such as the striations on the ribs and pectoral muscles. These are the kinds of microstructures that can quickly turn a typical smoothing pass into sports gruel.
In Adonis 2.1, AdnMush runs within every solver as part of the solver itself, so the output requires less post-processing. It is also available as a standalone post-processing node for shot-level corrections when artists need to clean up the results without re-running the full simulation. For users of the Turbo Auto Setup workflow, the generated rigs now use AdnMush for artifact fixing, reducing the amount of cleanup required before the setup is ready for initial supervisor feedback.
price and compatibility
Adonis 2.1 is available now. At Inbibo, AdonisFX Pro licenses start from £1,500 per year, with Indie licenses also available. AdonisML Pro licenses start from £3,000 per year and include AdonisFX, plus ML Deformer, SmartTissue, and associated machine learning tools. AdonisFX has a 30-day free trial, but you must contact Inbibo directly for evaluation access to AdonisML.
Adonis 2.1 is compatible with Maya 2023 and later and Houdini 20 and later. Supported operating systems are listed as Windows 10 and above, CentOS 7 and above, RHEL 8 and above, and Rocky Linux 8 and above. AdonisFX Indie subscriptions cost £25 per month or £250 per year per node-locked license for artists earning less than £100,000 per year. Pro subscriptions cost £110 per month or £1,100 per year for a node-locked license and £150 per month or £1,500 per year for a floating license.
Inbibo: Adonis 2.1 released
Download / Trial // Inbibo Download
