Simple, Transparent, Pay-As-You-Go
Only pay for what you use — every request converts directly into credits you can spend on any model in the mesh.
Add a card once, start building immediately.
No minimums, no monthly plans, just credits that never expire.
How It Works
Add Credits
Load your account with USD. Resonatia converts each dollar into Resonatia Credits you can spend across all models.
1 credit ≈ $1 USD
Use Any Model
Every call deducts a tiny amount of credits based on what you run — tokens, images, audio seconds, or video frames.
Pricing scales automatically by model size and quantization level (INT8 → Q4 → Q8).
Scale Freely
No locked tiers. Start small, burst when you need to.
Credits update in real time so you can track spend and performance.
Example Rates
Actual pricing varies slightly by model and region latency.
| Task Type | Example Model | Unit | Avg Cost |
|---|---|---|---|
| Chat / LLM | mesh/llama-3.1-q4 | per 1K tokens | $0.002 – $0.01 |
| Image Gen | mesh/sdxl-int8 | per image (1024×1024) | $0.02 – $0.10 |
| Audio | mesh/whisper-int8 | per second (STT / TTS) | $0.001 – $0.005 |
| Video | mesh/veo3-int8 | per frame (HD) | $0.0002 – $0.001 |
| Embeddings | mesh/e5-int8 | per 1K tokens | $0.001 |
Why It Feels Different
Edge-Optimized
Quantized models (INT8–Q8) run efficiently on phones, tablets, and local GPUs.
Rewarding Mesh
Every request is powered by independent nodes; your usage helps support the network.
Instant Access
No waitlists or approvals. Load credits, get your key, start sending requests.
Your Balance
Developer Console snippet
You pay in USD → instantly receive Resonatia Credits.
Credits are consumed per request and refill automatically if you enable auto-recharge.
Ready to build?
Create an account, add your first credits, and start quantized inference on the mesh today.