Oryx-34B-Image / README.md
THUdyh's picture
Update README.md
872e038 verified
|
raw
history blame
943 Bytes
metadata
license: apache-2.0
datasets:
  - THUdyh/Oryx-Image-Data
base_model:
  - Qwen/Qwen2-7B-Instruct
pipeline_tag: text-generation

Oryx-34B-Image

Model Summary

The Oryx-Image models are 7/34B parameter models trained on Oryx-Image-Data, based on Qwen2 language model with a context window of 32K tokens.

Oryx offers an on-demand solution to seamlessly and efficiently process visual inputs with arbitrary spatial sizes and temporal lengths.

Model Architecture

  • Architecture: Pre-trained Oryx-ViT + Yi-1.5-34B
  • Data: a mixture of 4M image data
  • Precision: BFloat16

Hardware & Software

  • Hardware: 64 * NVIDIA Tesla A100
  • Orchestration: HuggingFace Trainer
  • Code: Pytorch

Citation