| SenseNovaVision7b weights |
| ------------------------- |
|
|
| This repository is a byte-identical mirror of |
| sensenova/SenseNova-Vision-7B-MoT |
| (https://huggingface.co/sensenova/SenseNova-Vision-7B-MoT), |
| revision 79548fcc5b954598799b9317f8d3ec5e347d5c0e. |
| Copyright (c) 2026 SenseTime Group Inc. and/or its affiliates. |
| The model weights are licensed under CC BY-NC 4.0 (non-commercial). |
|
|
| The model architecture and reference implementation come from |
| OpenSenseNova/SenseNova-Vision |
| (https://github.com/OpenSenseNova/SenseNova-Vision), Apache License 2.0, |
| which builds on: |
|
|
| * Bagel, Copyright (c) 2025 Bytedance Ltd. and/or its affiliates, |
| Apache License 2.0 (https://github.com/ByteDance-Seed/Bagel) |
| * Qwen2 and SigLIP modules from Hugging Face transformers, |
| Copyright (c) 2024 The Qwen Team and The HuggingFace Inc. team, |
| Apache License 2.0 (https://github.com/huggingface/transformers) |
| * The FLUX autoencoder, Copyright (c) 2024 Black Forest Labs, |
| Apache License 2.0 (https://github.com/black-forest-labs/flux) |
|
|
| Paper: "Vision as Unified Multimodal Generation", arXiv:2607.06560. |
|
|
| LibreYOLO (https://github.com/LibreYOLO/libreyolo) consumes these weights |
| through its LibreVLM tier and prints the non-commercial license notice |
| before every automatic download. |
|
|
