What is Genius?

The OpenServe Genius model

OpenServe Genius is a Company-branded model, not a separate foundation model: it consists of OpenServe-configured routing, prompting, safety and orchestration over unmodified third-party models accessed by API. Genius currently routes each request to any one of the engines in the Third-Party Models & AI Notice depending on the task. OpenServe does not fine-tune, retrain or host the underlying models, and the engines behind Genius may change as quality and availability evolve; the current engine list is maintained on that notice.

The models behind Genius

Genius routes each request to the model best suited to it, from frontier models built by third-party AI labs to models we configure and operate ourselves. When you pick a specific model in the model picker, that model answers; when you pick Genius, the routing is automatic. The matrix below lists every third-party model maker across OpenServe — the chat models Genius can route to, the image and video models in Studio, and the supporting AI services — with each maker's own terms and privacy commitments.

Links verified August 28, 2026; makers revise these documents from time to time, and the model lineup changes as quality and availability evolve. Open-weight models (such as the Llama, Qwen, GLM, DeepSeek and GPT-OSS families) are run for us by hosting providers reached through Vercel AI Gateway, so both the maker's license and the hosting provider's terms can apply. The governing schedule of models, providers and routes is the Third-Party Models & AI Notice.