---
license: apache-2.0
---
# Model Card for cerebras/Cerebras-LLaVA-7B

The checkpoints consists of Language encoder and projector weights of multimodal LLaVA-7B model trained with our Cerebras implementation and training recipe. 
The vision encoder checkpoints for this model can be found at [cerebras/Cerebras-ViT-L-336-patch14-llava7b-ShareGPT4V](https://huggingface.co/cerebras/Cerebras-ViT-L-336-patch14-llava7b-ShareGPT4V)

**Note**: _ShareGPT4V_ is added to the vision model name to ensure correct loading of checkpoints in [LLaVA source repo](https://github.com/haotian-liu/LLaVA/blob/main/llava/model/multimodal_encoder/builder.py#L8)

For full details of this model and training details, please read our paper and release blog post **to be released shortly**.

# Model Architecture
Cerebras-LLaVA-7B is a transformer model with the following architecture details
* Vision encoder: [CLIP-VisionModel-Large](cerebras/Cerebras-ViT-L-336-patch14-llava7b-ShareGPT4V). It handles images of size 336 x 336 with patch size of 14
* Large Language Model: Pretrained from Vicuna-7B checkpoints and instruction finetuned on various datasets.
* Projector: the projector module that connects the LLM and Vision encoder part consists of two linear layers with gelu activation (mlp2x-gelu)

# Loading the model

This model can directly be loaded using the [LLaVa source code repository](https://github.com/haotian-liu/LLaVA). For installation, please refer to the [instructions in source code repository](https://github.com/haotian-liu/LLaVA?tab=readme-ov-file#install).

```
from llava.model.builder import load_pretrained_model
from llava.mm_utils import get_model_name_from_path
from llava.eval.run_llava import eval_model

model_path = "cerebras/Cerebras-LLaVA-7B"

tokenizer, model, image_processor, context_len = load_pretrained_model(
    model_path=model_path,
    model_base=None,
    model_name=get_model_name_from_path(model_path)
)
```

# Acknowledgements
We are thankful to all Cerebras engineers, past and present, that made this work possible.