116410271573381954849Q: LLM models and BYOK options
Since this offer landed on AppSumo a few days ago, I've been considering it. I think this could add a lot to my workflow and I very much appreciate being able to add LTDs products as supplements to some of my tools that represent ongoing costs.
Going through the website and other articles I could find on Noem.ai, I was left with two questions.
1.) I was curious which LLM (ChatGPT, Claude, DeepSeek) are the AI agents built on? I could not find any reference to that anywhere. And as I'm building out my stack of different softwares, this is of interest to me.
2.) Is there the option to BYOK? I'm assuming the answer is "no", since AI usage seems to be about internal credit with tokens, and there's no reference to it that I could find on that. But I did want to clarify.
Thank you, and I look forward to the response.
Daniel_NoemAI
Aug 3, 2025A: Thanks for the thoughtful questions, and we're glad you're considering adding Noem.ai to your workflow. Here's some clarity on both points:
1. Noem.ai currently uses a mix of large language models, primarily OpenAI and Claude, with others integrated as well. Rather than locking into a single model, we dynamically route requests to the best-performing model for the task. For example, real-time chatbots prioritize low-latency models, while reasoning-heavy tasks may use more advanced agents. This flexibility allows us to adapt as new models become available and as performance benchmarks shift, which is why we don’t publicly list specific models per use case. This all happens behind the scene so you never have to worry about it.
2. “Bring Your Own Knowledge,” yes, that’s fully supported and engouraged. You can import your own knowledge into the system via documents, text, URLs, or APIs and apply it to your private chatbot or AI Agent. However, if you meant “Bring Your Own Key” for running your own LLM billing (such as using your own OpenAI API key), that is not currently supported. AI usage is handled through our internal token system. We have enterprise licenses with our AI vendors since we pool our usage. This gives you access to the best models, on the best plans for a cheaper price.