https://www.submissionaudio.com/products/prefire

Bring your tracks to life with the rich analogue tone of 11 legendary microphone preamps, hand-captured with next generation neural network technology.

$99 USD
Save $50 USD

Introducing: PreFire
For over 30 years, engineers and producers have sought the analogue magic lost when studios transitioned from pristine, large-format analogue consoles to digital workflows. While digital was inevitable, early plugins often missed the essential non-linearities and harmonic saturation that defined classic gear. Even today's best models frequently compromise accuracy to save CPU power.
PreFire changes everything.
Utilizing advanced neural networks, PreFire captures every nuance and unique sonic detail of 11 legendary preamps across their entire gain ranges. From subtle enhancements of depth, width, and thickness, to rich saturation and extreme distortion, PreFire provides the authentic analogue character that defined countless timeless records.
PreFire is the ultimate Swiss army knife for your productions.
The Preamps
PreFire includes a suite of 11 legendary studio preamps, hand-selected for their unique tone and ability to bring something special to your music.

 

Official collab between Submission Audio (Flatline) and Neural Amp Modeler.

 

I just tried the demo and it seems quite interesting. These use the parametric Neural Amp Models,

https://www.neuralampmodeler.com/po...efire-powered-by-parametric-neural-amp-models.

 

Last edited: Dec 12, 2025

I was wondering when we were going to start seeing some parametric NAM models. Even though it has been available for something like 2 years, the only plugin we have seen so far is the Parametric OD that Steven Atkinson made just as an example of how you can make parametric NAM models. I thought it was going to catch on immediately, but I also have no idea how complex it is to make them. Lets hope that more developers see the value in using NAM parametric models for capturing frequency response and saturation of analog units in high detail.

I also really hope that Steven figures out some way to capture time domain effects like compression with NAM!

 

I think this is the future, and it's less clunky than Nebula, although the latter is still worth persevering with.

PS Steven is now working full-time on NAM and he is available to assist developers. The only problem is his technology negates the legacy algo emulations, which = $$$$$$ to the established brands.

 

Last edited: Dec 12, 2025

Good to know that he can work fulltime with it! Not many open-source developers can do that.

Pretty much all other brands are also going this way, either by using NAM or their own proprietary technology, so I doubt that big established brands are sweating it over NAM. It's probably more likely that it has lit a fire under their asses to get a foot in the game themselves. NAM just happened to be the first free and open-source player in the black-box modelling game. Kemper, Overloud, Neural DSP and IK Multimedia were all before NAM.