Back to Documentation
⚡
Spark
Fastest & most cost-efficient tier.
Intelligence
🧠
Basic
Reasoning
💡
Light
Speed
⚡
Very Fast
Input/1M
📥
$0.18
Output/1M
📤
$0.65
Spark is Ravonix’s fastest and most cost-efficient model. It is optimized for high-volume, low-latency workloads where responsiveness and scale are more important than deep reasoning.
While Spark excels in its primary domains, it is capable of performing across all domains; however, optimal results are most guaranteed when used as recommended.
Specifications
- Input Token Limit1,048,576
- Output Token Limit65,536
- ProfileHigh Efficiency
Regional Limits
| Plan | RPM | TPM |
|---|---|---|
| Starter | 20 | 40k |
| Basic | 50 | 100k |
| Pro | 150 | 250k |
| Enterprise | 400 | 500k |
| Custom | Custom | Custom |