ZMIME

Hy4 Base vs Shot Reader 8B

Compare documented capabilities and access. A dash means a comparable figure is unavailable.

Model specifications
SpecificationHy4 BaseTencentShot Reader 8BVoyager
TaskMultimodal pretrainingShot analysis
Published capabilitiesPretrained checkpoint for continued training and adaptation. Tencent’s flagship Hy4 uses a 78-layer language backbone with a separate vision encoder.Extracts camera framing, lens, movement and subject positions from footage using a fine-tuned Qwen3-VL model.
Model size770B total / 49B active per token in the language backbone8B backbone
AccessDownloadable under the Tencent Hy LicenseDownloadable weights
Result—Structured JSON shot sheet
SourcesModel weightsModel weights
InputsText, Images, VideoText, Images, Video
WeightsOpen weights · Tencent Hy LicenseOpen weights · Apache-2.0
ReleasedNov 20, 2026Nov 19, 2026