lmms-lab
/

LLaVA-OneVision-1.5-4B-Instruct

Image-Text-to-Text

feature-extraction

Model card Files Files and versions

Metrics Training metrics Community

xiangan commited on Oct 5

Commit

6430d15

·

verified ·

1 Parent(s): 967d049

Update README.md

Files changed (1) hide show

README.md +0 -1

README.md CHANGED Viewed

@@ -31,7 +31,6 @@ This is the official Hugging Face model card for **LLaVA-OneVision-1.5**, a nove
 3.  **Ultra-Efficient Training Framework**
     Complete end-to-end training framework designed for maximum efficiency:
     - **$16K total budget** for full model training
-    - **45% HFU efficiency** on A100 GPUs ($0.6 per GPU/Hour)
     - Built on **MegatronLM** with support for **MoE**, **FP8**, and **long sequence parallelization**
     - Optimized codebase for cost-effective scaling

 3.  **Ultra-Efficient Training Framework**
     Complete end-to-end training framework designed for maximum efficiency:
     - **$16K total budget** for full model training
     - Built on **MegatronLM** with support for **MoE**, **FP8**, and **long sequence parallelization**
     - Optimized codebase for cost-effective scaling