Diffio API models

Model cards for audio restoration

Compare speed, fidelity, and pricing, then pick the model id that fits your workflow.

View pricing

Model cards

Each card includes the model id, best use cases, and the current API rate.

Diffio 2.0 Flash

Fastest turnaround for previews and quick cleanups.

Fast
Model iddiffio-2-flash
Best forRapid previews, high volume queues.
Language supportMultilingual
  • Low latency option for high volume queues.
  • Great for rapid iterations and early reviews.
  • Balanced cleanup for shorter clips.
Price per secondLoading pricing
Estimated per hourLoading pricing

Billed by processed seconds, invoiced monthly. Minimum billable duration is 60 seconds per generation request.

View billing details

Diffio 2.0

Reliable baseline model for consistent cleanup.

Balanced
Model iddiffio-2
Best forProduction cleanup with balanced speed and fidelity.
Language supportMultilingual
  • Steady results for interviews and podcasts.
  • Good balance of speed and quality.
  • Stable output across large batches.
Price per secondLoading pricing
Estimated per hourLoading pricing

Billed by processed seconds, invoiced monthly. Minimum billable duration is 60 seconds per generation request.

View billing details

Diffio 3.4

Best quality reconstruction for demanding speech.

Best
Model iddiffio-3.4
Best forCritical clarity work and detailed transcripts.
Language supportEnglish only
  • Improved clarity in noisy rooms.
  • Sharper consonants and tighter detail.
  • Stronger alignment for transcript review.
Price per secondLoading pricing
Estimated per hourLoading pricing

Billed by processed seconds, invoiced monthly. Minimum billable duration is 60 seconds per generation request.

View billing details

Get started

Ship your first restoration quickly

Generate an API key, follow the docs, and run a request in minutes.

View documentation