FastH3-Omni-MLX-INT4 / root /text_encoder /preprocessor_config.json
Disrupt-linus's picture
Add complete Qwen-VL text encoder weights and tokenizers for 1-shot self-contained inference
7b4e9b0 verified
Raw History Blame Contribute Delete
390 Bytes
{
"size": {
"longest_edge": 16777216,
"shortest_edge": 65536
},
"patch_size": 16,
"temporal_patch_size": 2,
"merge_size": 2,
"image_mean": [
0.5,
0.5,
0.5
],
"image_std": [
0.5,
0.5,
0.5
],
"processor_class": "Qwen3VLProcessor",
"image_processor_type": "Qwen2VLImageProcessorFast"
}