Instructions to use meta-models/Muse-Glimmer-30B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use meta-models/Muse-Glimmer-30B with Transformers:

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

pipe = pipeline("image-text-to-text", model="meta-models/Muse-Glimmer-30B")
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("meta-models/Muse-Glimmer-30B")
model = AutoModelForMultimodalLM.from_pretrained("meta-models/Muse-Glimmer-30B", 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 meta-models/Muse-Glimmer-30B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "meta-models/Muse-Glimmer-30B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meta-models/Muse-Glimmer-30B",
		"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/meta-models/Muse-Glimmer-30B

SGLang

How to use meta-models/Muse-Glimmer-30B 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 "meta-models/Muse-Glimmer-30B" \
    --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": "meta-models/Muse-Glimmer-30B",
		"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 "meta-models/Muse-Glimmer-30B" \
        --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": "meta-models/Muse-Glimmer-30B",
		"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 meta-models/Muse-Glimmer-30B with Docker Model Runner:
```
docker model run hf.co/meta-models/Muse-Glimmer-30B
```

F* U LIZARDBERG

#32

by icedmatchalatt3 - opened about 7 hours ago

Discussion

icedmatchalatt3

about 7 hours ago

G@yest shit ever,although i shouldn’t have expected anything else from this hypocritical lizard..
It’s insane that we (Chinese) still keep getting discriminated against in every possible way while you all get unrestricted access to our latest models.I hope our labs one day grow some balls and impose geo-restrictions,even though i understand that such bitch ass,f@ggoty tendencies are the exclusive trademark of american corporations..
You're all just a bunch of pathetic,two-faced scumbags.Take your worthless piece of crap model and stick it so far up your ass you choke on it.
Qwen 3.8 will wipe the floor with it anyway..

nnnig

about 6 hours ago

stfu bro
have you heard of a vpn? its even on modelscope??
im dont like american corporations either but you sound stupid with that attitude and making everyone else look bad its really not helping

UniversalLove333

about 6 hours ago

You can use a VPN... or go to another source on HF...
You'll get to try it.
And yea prob Owen might be better, who knows... currently.

gman402

about 5 hours ago

Chinamen don't have access to VPNs, its illegal in their shithole country lol

jokerface1217

about 5 hours ago

good job

icedmatchalatt3

about 5 hours ago

•

edited about 4 hours ago

have you heard of a vpn?

How is the fucking idea of "Use a VPN to access an open weight model" even acceptable?And frankly,why should i even treat this as something acceptable when i'm being selectively shut out?What kind of segregation is this and why is this considered completely normal?Or it's just simply not as open as they make it out to be...

icedmatchalatt3

about 5 hours ago

its illegal in their shithole country lol

Yeah,there's always some american garbage in the comment section who thinks he’s the big fish in the pond.
Keep your VPN tips for Utah...

nnnig

about 4 hours ago

How is the fucking idea of "Use a VPN to access an open weight model" even acceptable?And frankly,why should i even treat this as something acceptable when i'm being selectively shut out?What kind of segregation is this and why is this considered completely normal?Or it's just simply not as open as they make it out to be...

you need to relax. i never said using a vpn to access an open weight model is acceptable. im pointing out the fact that you crying and whiny like a baby isnt changing anything and giving you country access to their models.
there is literally a modelscope mirror of this exact model by meta themselves, why dont you go download from there?

NodeLinker

about 4 hours ago

Yeah, the truth is, this has never been the case at Meta before, only now, huh?) They really do restrict access for users from US-sanctioned countries more harshly than anyone else, even compared to other American companies. Maybe it's personal animosity, who knows? Still, Meta is probably just being overly cautious. Nothing's stopping you from bypassing this fence, which doesn't actually exist anyway.

Ashacorporation

30 minutes ago

Wait, didn't China ban Hugging Face back in 2023? So using a VPN to bypass your own government's firewall is totally fine, but using a VPN to bypass Meta's restriction is where you draw the line? Make it make sense

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