Artificial Intelligence
      Back to home
      1. Innoslate Help Center
      2. Integrations
      3. Artificial Intelligence

      AI Models and Platform Options

      An overview of the 4.11 AI models available to integrate with Innoslate.

      Overview of AI Integrations

      Innoslate provides integration capabilities with leading AI platforms such as OpenAI, AskSage, and Azure OpenAI. These integrations enable Innoslate users with advanced AI-driven tools and services designed to enhance their applications and workflows. It is important to note that the options for self-hosting or on-premises deployment differ across different AI platforms.

      Below, we present an overview of these integrations, along with the self-hosting possibilities for each:

      • Innoslate Cloud is enabled by default to utilize OpenAI's GPT-4o mini model. If you are overseeing your organization's Cloud environment and do not wish to utilize the AI Chat feature, you may choose to disable this setting. Additionally, the AI Image API Keys are turned off. 
        • Trial users and academic organizations using Innoslate Cloud have a token limit of 5000 tokens per day for all users.
      • Innoslate GovCloud is defaulted to off; when activated, it employs OpenAI's GPT-4o mini model. If you are overseeing your organization's GovCloud environment and wish to utilize the AI Chat feature, you may choose to enable this setting. Additionally, the AI Image API Keys are turned off.
      • Innoslate Enterprise is adaptable, configured according to the specific model requirements of your environment. The token limit per organization can be customized to meet your specific requirements.

      Please be advised that when utilizing alternative models, users must provide their own API Key. Token configurations can be adjusted to meet your specific requirements.

      OpenAI Integration

      OpenAI is one of the most widely recognized AI platforms. Innoslate integrates with OpenAI's models to provide AI-powered capabilities such as text generation, test and risk generation, and conversational agents.

      Self-Hosting with OpenAI: OpenAI cannot be self-hosted in the traditional sense. This is because OpenAI’s models, including GPT-4 and DallE, require API and endpoint access to function. All interactions with OpenAI are conducted via API calls to OpenAI’s cloud-based infrastructure. As a result, software vendors using Innoslate's OpenAI integration must rely on OpenAI’s cloud infrastructure for model access and deployment, and cannot host the models locally.

      ChatGPT Enterprise is an available solution that provides businesses with enhanced security, privacy, and performance while still leveraging OpenAI’s cloud-based infrastructure. ChatGPT Enterprise does not enable full self-hosting. Organizations must still rely on OpenAI’s servers to process requests, meaning all AI interactions remain dependent on OpenAI’s cloud environment. For enterprises that require on-premise AI solutions due to strict security or compliance requirements, alternative approaches such as AskSage may be necessary to work with Innoslate.

      Considerations:

      • API Dependency: The reliance on OpenAI’s API means that you need an internet connection to access and use the service.
      • Scalability: OpenAI offers scalable services through its API, allowing vendors to scale up or down depending on usage requirements
      • Data Privacy: Vendors need to consider data privacy and compliance regulations, as the data is processed in OpenAI’s cloud environment.
      • Pricing Options: OpenAI follows a pay-as-you-go model based on token usage. For businesses, ChatGPT Enterprise offers a subscription. They also provides volume discounts for high-usage customers and custom pricing for large-scale deployments.

      AskSage Integration

      AskSage is another powerful AI service that Innoslate integrates with to offer natural language understanding and generation capabilities. AskSage’s integration provides access to a wide range of AI-powered tools for various applications, from simple queries to more complex problem-solving tasks. AskSage provides users with access to a diverse range of AI models from various company deployments, including some that are CUI-compliant.

      Self-Hosting with AskSage: AskSage offers the ability to have dedicated on-premise deployment/tenant at a price. This means that vendors can deploy the AskSage platform on their own infrastructure, giving them greater control over the AI models and data processing. 

      Considerations:

      • Deployment Flexibility: AskSage’s self-hosting capability gives vendors full control over their deployment, allowing for on-premises or hybrid cloud architectures.
      • Personal Playground Canva: AskSage offers an AI workspace where users can experiment with different models, personas, digest files, test prompts, and fine-tune responses in a controlled environment.
      • Pricing Model: AskSage operates on a monthly subscription-based model, with tokenization varying based on the AI model used.

      Azure OpenAI Integration

      Azure OpenAI is a service provided by Microsoft, offering access to OpenAI's powerful models through the Azure platform. Innoslate integrates with Azure OpenAI models, allowing users to leverage the same advanced AI capabilities that OpenAI offers, but redirecting to the Azure ecosystem.

      Self-Hosting with Azure OpenAI: Like OpenAI, Azure OpenAI cannot be fully self-hosted. It requires API and endpoint access for interaction with models, meaning vendors must rely on Azure’s cloud infrastructure for deployment. However, Azure offers Azure Government Models in regions such as USGovArizona, USGovVirginia, and USGovTexas, supporting IL4 and IL5 security compliance for highly regulated industries.

      * Important: To apply to get access to the Azure Governments in your instance, complete this form.

      Considerations:

      • Cloud Integration: Azure OpenAI enables seamless integration with other Azure services, providing a broad ecosystem for cloud-native applications all under one/or multiple subscriptions.
      • Hybrid API Service: While full self-hosting is not available, Azure’s hybrid cloud options (such as Azure Arc) may provide some flexibility for vendors looking for a combination of cloud and on-premises capabilities.
      • Pricing Options: Azure OpenAI offers Standard (On-Demand) pay-as-you-go pricing, Provisioned Throughput Units (PTUs) for predictable costs with reserved capacity, and a Batch API for discounted, non-urgent completions within 24 hours. 

      Conclusion

      Innoslate’s integration with OpenAI, AskSage, and Azure OpenAI provides Innoslate users with powerful AI-driven tools, each offering different self-hosting capabilities:

      • OpenAI and Azure OpenAI require API access and rely on cloud infrastructure for deployment.
      • AskSage provides both API and self-hosting options, allowing vendors to deploy AI models on their own infrastructure.

      For vendors with strict regulatory, privacy, or operational requirements, AskSage presents a compelling self-hosting option. Meanwhile, OpenAI and Azure OpenAI offer advanced AI capabilities for those comfortable with cloud-based deployments.