SenseNova-Loop-9B-Preview

Use it for tasks such as document understanding, coding, and interaction with external tools.

Availability

Model weights on Hugging Face

Pricing model

Not established

Deployment

Transformers, LightLLM, and SGLang deployment instructions

Privacy

Not established

Licence

Apache License 2.0

• The model accepts text and images and generates text. ^[1]

  It accepts text and images and generates text for applications that combine document understanding, coding, and interaction with external tools.

• It is designed for long-context and agentic workflows. ^[1]

  - `[2026.10.30]` We release **SenseNova-Loop-9B-Preview**, a looped multimodal model for long-context and agentic workflows.

• Its training context is up to 262,144 tokens. ^[1]

  Documented training recipes combine long sequences, multimodal inputs, and supervised task trajectories, with a maximum training sequence length of 262,144 tokens.

• Model weights are available from the Hugging Face model page. ^[1]

  Use Transformers to load the model and its custom implementation from a Hugging Face repository or a local model directory.

• The model can be loaded with Transformers using its custom implementation. ^[1]

  Use Transformers to load the model and its custom implementation from a Hugging Face repository or a local model directory.

• The model is licensed under Apache License 2.0. ^[1]