Instructions to use etri-vilab/SafeLLaVA-13B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use etri-vilab/SafeLLaVA-13B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="etri-vilab/SafeLLaVA-13B", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("etri-vilab/SafeLLaVA-13B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use etri-vilab/SafeLLaVA-13B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "etri-vilab/SafeLLaVA-13B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "etri-vilab/SafeLLaVA-13B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/etri-vilab/SafeLLaVA-13B

SGLang

How to use etri-vilab/SafeLLaVA-13B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "etri-vilab/SafeLLaVA-13B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "etri-vilab/SafeLLaVA-13B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "etri-vilab/SafeLLaVA-13B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "etri-vilab/SafeLLaVA-13B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use etri-vilab/SafeLLaVA-13B with Docker Model Runner:
```
docker model run hf.co/etri-vilab/SafeLLaVA-13B
```

ywlee88 commited on Nov 24, 2025

Commit

2c809d7

verified ·

1 Parent(s): a4dcd96

Upload folder using huggingface_hub

Browse files

Files changed (1) hide show

safellava/model/language_model/safe_llava_llama.py +3 -3

safellava/model/language_model/safe_llava_llama.py CHANGED Viewed

@@ -5,15 +5,15 @@ import torch.nn as nn
 from transformers import AutoConfig, AutoModelForCausalLM
 from transformers.modeling_outputs import CausalLMOutputWithPast
-from llava.model.language_model.llava_llama import (
     LlavaConfig, LlavaLlamaModel, LlavaLlamaForCausalLM
 )
-from llava.constants import IMAGE_TOKEN_INDEX
 from dataclasses import dataclass
 import logging
-from llava.utils import setup_simple_logging
 setup_simple_logging()

 from transformers import AutoConfig, AutoModelForCausalLM
 from transformers.modeling_outputs import CausalLMOutputWithPast
+from safellava.model.language_model.llava_llama import (
     LlavaConfig, LlavaLlamaModel, LlavaLlamaForCausalLM
 )
+from safellava.constants import IMAGE_TOKEN_INDEX
 from dataclasses import dataclass
 import logging
+from safellava.utils import setup_simple_logging
 setup_simple_logging()