Instructions to use prithivMLmods/Holo3-35B-A3B-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use prithivMLmods/Holo3-35B-A3B-FP8 with Transformers:

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

pipe = pipeline("image-text-to-text", model="prithivMLmods/Holo3-35B-A3B-FP8")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("prithivMLmods/Holo3-35B-A3B-FP8")
model = AutoModelForMultimodalLM.from_pretrained("prithivMLmods/Holo3-35B-A3B-FP8", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use prithivMLmods/Holo3-35B-A3B-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prithivMLmods/Holo3-35B-A3B-FP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/Holo3-35B-A3B-FP8",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/prithivMLmods/Holo3-35B-A3B-FP8

SGLang

How to use prithivMLmods/Holo3-35B-A3B-FP8 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 "prithivMLmods/Holo3-35B-A3B-FP8" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/Holo3-35B-A3B-FP8",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "prithivMLmods/Holo3-35B-A3B-FP8" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/Holo3-35B-A3B-FP8",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use prithivMLmods/Holo3-35B-A3B-FP8 with Docker Model Runner:
```
docker model run hf.co/prithivMLmods/Holo3-35B-A3B-FP8
```

Deployment

by akzaidan - opened Jun 8

Discussion

akzaidan

Jun 8

•

edited Jun 8

Hello, how do you recommend deploying this model? I am running into a ton of deployment errors on HG Inference Endpoints.

here is my deployment config:

AWS GPU · Nvidia RTX PRO 6000 Blackwell · 2x GPUs · 192 GB
vllm/vllm-openai:v0.18.1
Container Arguments
--trust-remote-code --max-model-len 32768 --disable-custom-all-reduce --enforce-eager --gpu-memory-utilization 0.85

Every time I try and deploy it seems that I run out of memory and I get this error Engine core initialization failed. I also tried to use 4x GPU's with more than enough memory and same issue.

Also I had to update "tokenizer_class" to "Qwen2TokenizerFast"

I'd appreciate help with these questions too:

Do you recommend trying to quantize the Holo3 35B model down to INT4?
What do you think the best model to use is for application form filling? By "best" I mean its quick, cheap, and does reasonably well filling out a job application for example.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment