Shot Reader 8B vs Hy4 Base
Compare documented capabilities and access. A dash means a comparable figure is unavailable.
| Specification | ||
|---|---|---|
| Task | Shot analysis | Multimodal pretraining |
| Published capabilities | Extracts camera framing, lens, movement and subject positions from footage using a fine-tuned Qwen3-VL model. | Pretrained checkpoint for continued training and adaptation. Tencent’s flagship Hy4 uses a 78-layer language backbone with a separate vision encoder. |
| Model size | 8B backbone | 770B total / 49B active per token in the language backbone |
| Access | Downloadable weights | Downloadable under the Tencent Hy License |
| Result | Structured JSON shot sheet | — |
| Sources | Model weights | Model weights |
| Inputs | Text, Images, Video | Text, Images, Video |
| Weights | Open weights · Apache-2.0 | Open weights · Tencent Hy License |
| Released | Nov 19, 2026 | Nov 20, 2026 |