Explore Neural Amp Modeler, discover great amp tones on TONE3000, and download my hand-picked Neural Captures for Neural DSP Cortex processors.
Modern amp modeling gives guitarists access to an incredible range of sounds without the need to own a collection of physical amplifiers. Two technologies that make this possible are Neural Amp Modeler (NAM) and Neural DSP’s proprietary Neural Capture technology.
On this page, I’ll explain what NAM files are, introduce the main NAM architectures, share some of my favorite models from TONE3000, and give you access to my own Neural Captures for the Cortex ecosystem.
Whether you’re exploring NAM for the first time or looking for new sounds for your Nano Cortex, Quad Cortex mini, or Quad Cortex, you’ll find everything in one place.
A little AI touch: Even the featured image for this article was created with the help of AI — another example of how these technologies are becoming part of the creative process for musicians. 😅
The gear shown in the image is illustrative.
1. What Is Neural Amp Modeler (NAM)?
Neural Amp Modeler (NAM) is an open-source project that uses neural networks to reproduce the sound and response of guitar amplifiers, pedals, and other audio equipment.
A NAM model is stored in a .nam file. You can load it into compatible software or hardware and use it to process your guitar signal in real time.
One of NAM’s biggest advantages is its community-driven library. Guitarists and creators share thousands of models, covering everything from classic tube amplifiers and boutique gear to high-gain setups and overdrive pedals.
You can explore and download NAM models on TONE3000:
Explore NAM Models on TONE3000
2. NAM Architectures: A1 and A2
NAM models are available in different architectures and sizes. These affect their processing requirements and compatibility with NAM playback software.
A1 — The Original Architecture
The original NAM architecture is now referred to as A1. It includes several model configurations:
- A1 Standard
- A1 Lite
- A1 Feather
- A1 Nano
These configurations differ in model size and processing requirements. Smaller versions are designed for more limited hardware, while larger versions use more processing resources.
Existing A1 models remain supported by compatible NAM players.
A2 — The Newer Architecture
A2 is the newer generation of NAM architecture, introduced in 2026 and now used as the standard for new NAM models.
It offers two main model sizes:
- A2-Lite — a smaller, more lightweight version.
- A2-Full — a larger version with greater model capacity.
A single A2 model file can run in either A2-Full or A2-Lite mode. Compatible software can select the appropriate size automatically or allow you to choose it manually.
The main idea behind A2 is to make NAM models more efficient while maintaining high modeling accuracy.
If you’re using NAM software, make sure it’s up to date and supports the architecture of the model you’re trying to load.
3. Where Can You Use NAM Files?
NAM files can be used with compatible software plugins, standalone applications, and hardware devices that support NAM playback.
They are particularly useful for guitarists who want to experiment with different amplifier sounds without investing in expensive physical equipment.
However, NAM files and Neural DSP’s proprietary Neural Capture technology are two different formats.
Nano Cortex, Quad Cortex mini, and Quad Cortex do not load standard .nam files directly. They use Neural Captures, which are part of the Neural DSP Cortex ecosystem.
That’s why I’ve put together a separate collection for Cortex users.
4. My Favorite NAM Models from TONE3000
I’ve spent quite a bit of time exploring the NAM library and trying different amplifier models.
Below, you’ll find a selection of my personal favorites. These are models I’ve found interesting and useful for playing and recording.
You can listen to them directly on this page using the embedded TONE3000 players.

My NAM Selection
ENGL Fireball 25
Amp + Cab Model
ENGL Fireball 25 with a Soldano 2×12 cabinet
View on TONE3000 — ENGL Fireball 25
Mesa Boogie Dual Rectifier Rev F – Red Channel + OD808 – High Gain
Amp + Cab Model
AMP SETTINGS: M5 P1.5 B3 M5 T9 G6 Vintage, Diode O808 D0 T5 B7
CAB: Mesa 4×12 cabinet with UK-made Celestion Vintage 30s speakers
MICS: SM57, R121, U87
View on TONE3000 — Mesa Boogie Dual Rectifier
Blackstar Series One 100 MK II Head
Amp Head Model
The Blackstar Series One 100 MK II is a 100-watt tube head with four channels and six voices, covering everything from clean tones to high-gain sounds. The TONE3000 entry includes three NAM models with different OD1 and OD2 settings.
View on TONE3000 — Blackstar Series One
Diezel VH4 Channel III – High Gain
Amp + Cab Model
AMP SETTINGS: G5 V5 T6 M5 B5.5 D6 P6 V5
CAB: Mesa 4×12 cabinet with UK-made Celestion Vintage 30s speakers
MICS: R121, SM57, U87
I’ll update this section as I discover more models worth sharing.
5. My Neural Captures for Neural DSP Cortex
If you’re using a Nano Cortex, Quad Cortex mini or Quad Cortex, this is the section you’ll probably find most interesting.
I’ve prepared a selection of Neural Captures and uploaded them to Cortex Cloud. You can find them there and try them on your compatible Neural DSP device.
You can now explore these sounds directly on your Cortex device without needing to load .nam files.
I’ve made my collection available through my Guitarist 365 Days profile on Cortex Cloud.
The collection includes my personal selection of amplifier tones and sounds that I find particularly useful for playing and recording.
I’ll keep adding new captures as I discover more interesting models and tones worth sharing.

6. Compatible Devices
My Neural Captures are intended for the Neural DSP Cortex ecosystem, including:
- Neural DSP Nano Cortex
- Neural DSP Quad Cortex
- Neural DSP Quad Cortex mini
Cortex Cloud lets you browse and download Neural Captures compatible with your selected device.
Make sure you select the correct device in Cortex Cloud and keep its firmware up to date to access the latest features.
Remember that these Neural Captures are separate from the original NAM files. You cannot simply copy a .nam file to your Cortex processor and load it as a Neural Capture.
7. Frequently Asked Questions
Can I load NAM files directly into Nano Cortex, Quad Cortex mini or Quad Cortex?
No. These devices use Neural Captures rather than standard .nam files. The captures in my Cortex Cloud collection are intended for compatible Cortex devices.
What’s the difference between NAM and Neural Capture?
NAM is an open-source modeling format that uses .nam files and compatible NAM playback software. Neural Capture is Neural DSP’s own technology, designed for its Cortex ecosystem. They are separate formats and aren’t directly interchangeable.
Where can I download free NAM files?
You can explore a large community library of NAM models on TONE3000. Many models are available to download and try with compatible NAM software.
Where can I find your Neural Captures?
Visit my Guitarist 365 Days profile on Cortex Cloud using the link in the section above. You can browse and download the available captures for your compatible device.
Final Thoughts
One of the things I enjoy most about modern amp modeling is how easy it has become to discover new sounds.
NAM gives guitarists access to a huge community-driven collection of amplifier models, while Neural DSP Cortex processors offer a convenient way to use Neural Captures in a compact hardware setup.
I created this page to share some of the sounds I’ve discovered, save you time searching through thousands of models, and make my own Cortex-compatible captures available in one place.
Feel free to explore the collection, try different tones, and find something that inspires your playing.
Enjoy the sounds!
Andrew Nitsievsky
Guitarist 365 Days

