2.5
By Initial Audio
Generate MIDI and convert audio to MIDI.Locally on your computer.
Play live or drop a MIDI idea, then Re-Imagine a new melody. Or drag an audio loop onto the piano roll. Version 2.5 also runs as VST3 and AU. No cloud. No subscriptions.
Audio to MIDI.Now a plugin.
Version 2.5 is a free update. Drag a melodic audio loop onto the piano roll and Difro extracts MIDI — notes and velocity — in a fraction of a second. And the whole app now runs as VST3 and AU, so you can generate, convert, and play in sync with your DAW.
-
Audio to MIDI Drop an audio loop on the piano roll and Difro pulls out the melody as MIDI, including velocity. Built for melodic loops up to 16 bars — a full song is truncated to the first 16 bars.
-
VST3 and AU plugin The full Difro experience inside your DAW. Sync output, stream MIDI and audio in real time, or mute below Master gain and route MIDI only.
-
Key triggering In the plugin, MIDI notes start and stop playback, with a latch option so one hit holds the loop. Turn it off and incoming notes play the instrument loaded inside Difro.
-
Interactive Quick Start A new in-app tutorial walks first-time users through the UI and the basics of generating, converting, and dragging out MIDI.
-
A workspace you can reshape Drag section edges to give more room to the piano roll, packs, or mixer. Brightness and contrast sliders live in the title bar. The UI is snappier throughout.
-
Lighter, ready for Blackwell About 300MB smaller to download and run. Windows now supports NVIDIA Blackwell GPUs. Older Maxwell cards are no longer supported.
The full Difro experience.Inside your DAW.
Difro 2.5 ships as a VST3 and AU plugin — the same piano roll, packs, and generation, locked to your project tempo. Output MIDI, audio, or both in real time.
- Sync playback to your DAW and stream MIDI and audio in real time.
- Want MIDI only? Mute below Master gain and route Difro's MIDI to another track in your DAW.
- Key triggering — MIDI notes start and stop the loop, with a latch option so one hit holds it.
- Turn key triggering off and incoming notes play the instrument loaded inside Difro.
Five new packs.Every factory pack rebuilt.
Click a pack and Difro 2 starts generating from its MIDI. The whole library was overhauled for the new model — plus five brand new packs.
Pop
Radio-ready hooks and chord beds.
Trance
Long-form lifts and festival lines.
Luxury Keys
Lush piano and keys voicings.
Omega
Experimental retro-game melodies.
Platinum Factory
Trap and hip-hop factory heat.
See Difro 2.5in action.
The 2.5 trailer — Audio to MIDI, the VST3 / AU plugin, key triggering, and the rest of the update.
One click to unlimited new melodies.
No music theory degree required.
1. Open & Click
Difro 2 ships with a rebuilt library of MIDI Prompt Packs, plus five new ones. Open the app or the plugin, click a pack, and it starts generating.
2. Play, drop MIDI, or drop audio
Hit record and play into the piano roll, drag MIDI in, or drop an audio loop on the roll — Difro extracts the melody as MIDI, velocity included. Then Re-Imagine it.
3. Drag into your DAW — or stay in it
The new melody appears bar-by-bar in real time. Drag MIDI or a WAV loop into Ableton, FL Studio, Logic, or Cubase — or run Difro as VST3 / AU and output MIDI and audio in sync.
Not a randomizer. A real musician.
Difro actually understands melody, rhythm, and chord progressions. It keeps the DNA of your original idea while creating something entirely fresh.
ReImagine Any MIDI
No MIDI file required — hit record and play live, drag in chords, or drop an audio loop on the piano roll to extract MIDI. Then use the Creativity Slider to decide how far Difro should take it.
Host Your Favorite Plugins
Load Serum, Kontakt, or Omnisphere directly inside Difro. Audition your new AI melodies with your exact presets before dragging them into your DAW.
Shape it after you generate
Gate, Arp, Rhythm, and Melodic Direction apply after the notes arrive — extra variation without throwing the idea away. Arp now snaps timings to the grid.
CPU or GPU. GPU when you have it.
Difro 2.5 defaults to GPU when a compatible card is available, including NVIDIA Blackwell on Windows. No NVIDIA or Apple silicon? It stays in CPU mode — fully offline either way.
SourceLab LightBuilt Right In.
Every copy of Difro comes with a built-in Light Version of our flagship synth, SourceLab. You don't need to load external plugins to hear your creations. The player has its own UI — a simple browser and macros — while presets still work from the current menu.
- Includes 80+ high-quality SourceLab presets.
- New player UI with a simple browser and macros.
- Presets can still be selected from the current menu.
- Audition MIDI files instantly with premium sounds.
- Seamless integration inside the Difro standalone app.
Host Your FavoritePlugins.
Load Serum, Kontakt, or Omnisphere directly inside Difro — in the standalone app and in the new VST3 / AU plugin. Audition your new AI melodies with your exact presets before you export.
- VST3 plugin hosting on Windows
- Audio Unit (AU) hosting on macOS
- Preview with your own presets before you export
- Works in the standalone app and inside the Difro plugin
Bounce a loop.Drag it anywhere.
Hit Render to WAV and Difro bounces the generated melody to an audio loop. Drag the WAV to your desktop, a custom folder, or straight into your DAW as a sample. Prefer MIDI? Click the output MIDI clip right below the Render to WAV button and drag that out instead.
- Render the output MIDI to a WAV loop in one click.
- Drag the WAV to your desktop, a folder, or into your DAW as a sample.
- Dragged-out audio files now include Key/Scale and BPM in the file name.
- Drag MIDI out from the clip directly below the Render to WAV button.
An AI that adaptsto you.
Difro 2 is not a static model. Rate a generation and name the Feedback model — a personal adapter that learns the music you actually like. Keep several named models, pick Difro 1 or Difro 2 as the base, and switch anytime. Edit the MIDI first, then rate it positive, and the adapter learns how you expect the output to look.
- Rate any output with the happy or sad face to teach Difro what you like — and what you don't.
- Name each Feedback model so you can keep several personal adapters and switch between them.
- Edit the generated MIDI in the piano roll, then rate it with the happy face so the model learns your version as the target.
- Generate with the Difro 1 model, the Difro 2 model, or any named Feedback model.
- Session History keeps every generation and the original MIDI, ready to bookmark or reload.
Create CustomMIDI Prompt Packs.
Difro 2 ships with a rebuilt factory library and five new MIDI Prompt Packs. Want your own? Drop MIDI files into a folder — as many as you like, there's no limit — and Difro uses them as input prompts to generate variations in your exact style.
- Clicking a pack never repeats the same MIDI as the last pick.
- Right-click a pack card and choose Open model folder… to add or delete MIDI on your hard drive.
- Assign specific built-in SourceLab presets to your custom packs.
- 100% private — everything runs locally on your machine.
- Share your custom pack folders with the community.
Make It Yours. Multiple Skins Included.
The LLM ofMIDI.
Powered by a bleeding-edge combination of Variational Autoencoders (VAEs) and Transformer technology. True next-gen AI built directly into your local workflow.
Extracting the DNA
When you feed a MIDI clip into Difro, the VAE absorbs the soul of your idea, encoding it into a rich latent space - a multidimensional map of rhythm, contour, and harmonic movement.
Transformer Architecture
Just like an LLM predicts the next word, Difro predicts the perfect next note. It tracks long-range musical relationships, stringing together rhythms with human coherence.
Real-Time, Local Processing
Hyper-optimized to run locally on your CPU, with optional GPU acceleration. No cloud servers, no waiting. The neural network predicts notes so fast you hear the fresh melody stream directly.
CPU or GPU. Difro 2.5 defaults to GPU when a compatible NVIDIA card (Windows, including Blackwell) or Apple silicon GPU (Mac) is available. Older Maxwell GPUs on Windows are no longer supported. The 2.5 installer is about 300MB smaller. Switch anytime from the top-right corner. No GPU? It runs in CPU mode — no extra hardware required.
No Subscriptions. Just Inspiration.
Download the app or load the VST3 / AU plugin, convert audio to MIDI, and drag MIDI or a rendered WAV loop straight to your DAW.
Difro Free
Perfect to test the waters. Resets monthly.
- 80 Generations per month
- MIDI Prompt Packs (5 new + rebuilt library)
- Audio to MIDI + VST3 / AU plugin
- ReImagine & Transform Modes
- VST3 / AU Plugin Hosting
- Unlimited Generations
- Create Custom Models
Difro Pro
Black November — one-time payment. Yours forever.
- Unlimited Generations
- Create Custom Models
- MIDI Prompt Packs (5 new + rebuilt library)
- Audio to MIDI + VST3 / AU plugin
- ReImagine & Transform Modes
- 100% Royalty Free
- Lifetime Updates
Difro + SourceLab
Difro Pro and SourceLab 2 together. Re-Imagine MIDI in SourceLab.
- Difro Pro — unlimited generations
- SourceLab 2 + 4 free expansions
- Generate MIDI inside SourceLab
- 100% Royalty Free
- Lifetime Updates
Difro 2.5. Audio to MIDI. Inside your DAW.
Runs locally on your CPU or GPU. No cloud. No API bills. Ever.
The hard truth about "AI" music tools.Difro is the real thing.
A lot of so-called AI melody tools are not AI at all — the label sells, but under the hood it's mostly algorithms. Random note generators, rigid rule systems, or cheap cloud APIs with terrible output, strict limits, and fees that never stop. Difro is built differently: a real VAE + Transformer stack, running locally on your CPU or GPU, with no internet required and no ongoing costs.
What many tools actually are
- "AI" branding on top of basic randomization
- Requires a constant internet connection
- API usage fees and hard generation caps
- Output that ignores your chords and groove
- Your MIDI sent to third-party servers
What Difro actually runs
- Variational Autoencoder + Transformer neural net
- 100% offline — GPU by default when available
- One-time purchase, zero API bills
- Understands melody, rhythm & harmony
- Private — nothing leaves your machine
1. Extracting the DNA
When you feed MIDI into Difro, the VAE doesn't just read note data. It encodes the musical soul of your idea — rhythm, contour, and harmonic movement — into a rich latent space.
2. Transformer architecture
Just like ChatGPT predicts the next word, Difro's Transformer predicts the next note. It tracks long-range musical relationships and strings together phrases with human coherence.
3. Real-time, local inference
The entire model runs on your machine — GPU by default if you have NVIDIA on Windows (including Blackwell) or Apple silicon on Mac, otherwise CPU. No cloud queue, no upload step. You hear fresh melodies stream in real time, bar by bar.
To our knowledge, Difro is the world's first real MIDI GPT — a locally running generative AI that produces outstanding musical output without an internet connection or recurring costs. Think of it as a language model, but for MIDI.
