Neural Amp Modeler A2 Released

Steven Atkinson's NAM - Neural Amp Modeler , is a highly regarded piece of open-source software that uses deep learning to create models of guitar amplifiers and pedals with state-of-the-art accuracy. This new version 2 (A2) comes in two flavours and allows you ot train your own models using Tone3000s platform or using Stevens own Google Colab notebook. Much to love about a project so well realised, and made open source for the music making community. Here's more from Steve:

Neural Amp Modeler is an open-source project that uses deep learning to create models of guitar amplifiers and pedals with state-of-the-art accuracy. If you want a pre-made plugin that plays A2, you can check out Gateway over on the Users page.

To train your own A2 models, TONE3000 now supports training A2 models, as does my open-source Google Colab notebook or the local GUI-based trainer that ships with the training code.

If you want to build to support A2, then check out the following versions of the open-source code:

• neural-amp-modeler v0.13.0

• NeuralAmpModelerCore v0.5.2

• NeuralAmpModelerPlugin v0.7.14

I'll be filling out more on this blog post in the near future with the content from the video plus some more details that are better-suited for written-form content. For the moment, I'm spending a week off.

Here's some other coverage;

Support his work here

• MTV Jungle & DnB Music Videos 18-Sep-26
• Bitwig's New Native Devices 17-Sep-26
• Reason Scream4 is Now Free! 17-Sep-26
• Broke Beats pt1 - A £37 Interface & Multi FX 16-Sep-26
• Ed O'Brien's Must-Listen Albums 16-Sep-26