Neural Amp Modeler (NAM), by Steven Atkinson, is a free, open-source plugin that uses machine learning to recreate the exact tone of guitar amps, cabinets, and pedals — modeled from real hardware recordings instead of hand-tuned circuit simulations. Load one of the thousands of free community-made captures and it reproduces that gear’s dynamics, saturation, and character in real time, with a sound quality that regularly gets compared to modelers costing hundreds of dollars.
Features
- Real-time neural network processing of captured amp and pedal tones, preserving dynamic response and saturation.
- Gain and EQ controls, plus input/output level adjustment.
- Built-in IR loader for cabinet simulation.
- Works standalone or as a plugin inside your DAW.
- Thousands of free amp and pedal captures (.nam files) shared by the community on sites like ToneHunt.
- Train your own captures from a reamped signal using the free, open-source NAM trainer.
Compatibility
Available in VST3 and AU for Windows & macOS, plus LV2 for Linux. Free and open source, released under the MIT license.
Good to know
NAM itself doesn’t come with built-in amp presets — it’s a capture format, so sound quality depends on the .nam files you load or train yourself. High-resolution captures and IRs can be CPU-intensive on older machines.
Sources :
neuralampmodeler.com/: https://www.neuralampmodeler.com/users
NAM home: https://www.neuralampmodeler.com/
library: https://www.tone3000.com/
