Muse Glimmer ViT Runner#

class MuseGlimmerViTRunner : public trt_edgellm::rt::QwenViTRunner#

Muse-Glimmer vision encoder: window + full attention with a 2D interleaved rotary and a fast learned position embedding, run in raster token order (spatial_merge_size == 1); the merge_size == 2 pixel-shuffle is folded into reverse_window_index at the output.

Public Functions

QwenViTRunner(
std::string const &engineDir,
int32_t llmMaxBatchSize,
int32_t llmMaxSequenceLength,
cudaStream_t stream
)#

Constructor for QwenViTRunner.

Parameters:

  • engineDir – [in] Directory containing the TensorRT engine files

  • llmMaxBatchSize – [in] Maximum batch size from LLM engine

  • llmMaxSequenceLength – [in] Maximum sequence length from LLM engine

  • stream – [in] CUDA stream for execution

Throws:

  • std::runtime_error – if engine directory does not contain engine files, or if buffer allocation fails

  • json::type_error – if JSON configuration contains unexpected datatypes