FAQ
Frequently Asked Questions
Why Influxion?
Differences in AI model capabilities and hosting platforms are challenging to analyze. They are subject to evolving workloads, computational resource contention, pricing variability, and ever-changing business requirements. It is also becoming obvious that there is no single “best” model or platform, especially when considering dynamic, multidimensional behaviors like accuracy, latency, cost, and reliability.
Influxion is an Adaptive AI Orchestration Platform for harnessing AI. With Model Sets, you specify your requirements and we intelligently route your requests to model deployments for you. We handle all the complexities around infrastucture, monitoring, and fault tolerance so you can focus on what really matters—your application.
Getting Started
See Quickstart.
Models
Influxion provides a centralized Model Gateway that enables you to use a wide variety of model providers. Available models can be found on the Models page.
We continually add new models, but if you don't see what you're looking for, there's no need to wait—e-mail us at support@influxion.io and let us know.
Model Sets
Model Sets are a core Influxion feature that help you get the most out of LLMs.
You might think of a model set as a virtual model. Simply configure a model set with the behaviors you want it to exhibit, select which models may be used, and then use it just like it was a regular model. Influxion will intelligently route your request to the most desirable model endpoint to satisfy your custom requirements.
Pricing
Influxion charges 5% fee on top of model provider costs. For example, if you spend $1 per million tokens on a provider, we charge you $1.05.
If you bring your own provider API key, we charge only the 5% fee. For example, if you spend $1 per million tokens on a provider, we charge you $0.05.
All pricing is in U.S. Dollars (USD).
Rate Limits
Like most services, we enforce rate limits to ensure fair access to all users. At this time, rate limits are 1,000 request per minute (RPM) and 100,000 requests per day (RPD). We continue to tune these rate limits based on system demand and provider availability.
When you reach a rate limit, the gateway will return a 429 Too Many Requests status code. You will receive a similar response if you do not have enough credits in your account.
Downstream providers may also rate limit requests.
Privacy and Security
Influxion logs request metadata like latency and throughput metrics, which are essential to Model Set functionality. We may store request/response data for up to 30 days, which is used solely for our own system analysis and debugging.
Downstream model providers have their own data retention policies. We do not attempt to enumerate these at this time—please visit their documentation for details.
We encourage all users to apply good privacy practices. You are responsible for protecting your data from unauthorized access.
Terms of Service
View our Terms of Service.
Contact Us
We love to hear from our customers! Reach out to us if you have questions, encounter problems, or have other feedback.
Support: support@influxion.io