JSON2VIDEO Documentation Credit consumption
Credits are the unit of usage on JSON2Video. New accounts receive 600 free credits.
What credits are consumed for
- Rendering videos.
- Text-to-speech voiceovers.
Rendering
| Output resolution | Credits per second of output |
|---|---|
| Any resolution (SD up to 4K) | 1 |
The rendering cost depends only on the output duration — the resolution does not change the per-second rate.
Failed renders are not billed.
Voice assets
At the moment, the text-to-speech voiceovers are provided included in all plans. When the voice element specifies a connection ID, the upstream provider (Microsoft, ElevenLabs, …) bills the customer directly.
| Asset type | Model | Provider | Credits / unit |
|---|---|---|---|
| Voice | azure |
Microsoft | 0 per minute |
| Voice | elevenlabs-flash-v2-5 |
ElevenLabs | 0 per minute |
| Voice | elevenlabs-v3 |
ElevenLabs | 0 per minute |
| Subtitles | default / whisper |
JSON2Video | 0 per minute |
Cached assets are not re-billed. If the same input has been generated before (and the element's cache flag is true, which is the default), the cached version is reused.
Order of deduction
Credits are consumed in this order:
- Subscription-plan credits, earliest-expiring first.
- Prepaid (non-expiring) credits.
When neither has remaining credits, further renders fail with Insufficient credits. See Errors.
Worked examples
Simple Full HD video, no voiceover
1920×1080 × 10 seconds, no generated assets.
Cost: 10 credits.
Vertical video with a managed voiceover
1080×1920 × 30 seconds with a 20-second ElevenLabs voiceover using JSON2Video's managed key.
Cost: 30 (rendering) + 0 (ElevenLabs voice, currently included) = 30 credits. Voice models are included at 0 credits today; this may change as new models ship.
Same video, customer-supplied keys
1080×1920 × 30 seconds, voiceover routed through the customer's own ElevenLabs account via a Dashboard connection.
Cost: 30 credits (rendering only). The voice charge is billed directly by ElevenLabs.