From Dorbit

News

New models, product updates, and what’s next.

MiniCPM-V 4.7 is now supported

Run MiniCPM-V 4.7 35B-A3B and 1B on a dedicated endpoint.

Dorbit now supports MiniCPM-V 4.7 35B-A3B and MiniCPM-V 4.7 1B for inference. Both models support text, image, and video inputs and are available in the model catalog.

To get started, choose either model in Create Endpoint, compare compatible GPU servers, and select a duration. Dorbit takes care of the model and inference engine setup. Once your endpoint is ready, connect your application using its endpoint URL and the OpenAI-compatible API.

Strata is now supported

Run Qwen3.8-Flash-Next GGUF models on a dedicated endpoint.

Dorbit now supports Strata as an inference engine, starting with the IQ2_XS and IQ3_S GGUF variants of Qwen3.8-Flash-Next. Both are available in the model catalog and can be launched on a dedicated endpoint through Dorbit.

To get started, choose either variant in Create Endpoint, compare compatible GPU servers, and select a duration. Dorbit takes care of the inference engine setup. Once your endpoint is ready, connect your application using its endpoint URL and the OpenAI-compatible API.