Image-Text-to-Text
Transformers
Safetensors
muse_glimmer
compressed-tensors
int8
autoround
conversational
8-bit precision
auto-round
Instructions to use Minachist/Muse-Glimmer-30B-INT8-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Minachist/Muse-Glimmer-30B-INT8-AutoRound with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Minachist/Muse-Glimmer-30B-INT8-AutoRound") 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("Minachist/Muse-Glimmer-30B-INT8-AutoRound") model = AutoModelForMultimodalLM.from_pretrained("Minachist/Muse-Glimmer-30B-INT8-AutoRound", 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 Minachist/Muse-Glimmer-30B-INT8-AutoRound with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Minachist/Muse-Glimmer-30B-INT8-AutoRound" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Minachist/Muse-Glimmer-30B-INT8-AutoRound", "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/Minachist/Muse-Glimmer-30B-INT8-AutoRound
- SGLang
How to use Minachist/Muse-Glimmer-30B-INT8-AutoRound 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 "Minachist/Muse-Glimmer-30B-INT8-AutoRound" \ --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": "Minachist/Muse-Glimmer-30B-INT8-AutoRound", "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 "Minachist/Muse-Glimmer-30B-INT8-AutoRound" \ --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": "Minachist/Muse-Glimmer-30B-INT8-AutoRound", "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 Minachist/Muse-Glimmer-30B-INT8-AutoRound with Docker Model Runner:
docker model run hf.co/Minachist/Muse-Glimmer-30B-INT8-AutoRound
| { | |
| "bits": 8, | |
| "data_type": "int", | |
| "group_size": -1, | |
| "sym": true, | |
| "gradient_accumulate_steps": 2, | |
| "iters": 250, | |
| "low_gpu_mem_usage": true, | |
| "autoround_version": "0.15.0", | |
| "block_name_to_quantize": "model.language_model.layers", | |
| "quant_method": "auto-round", | |
| "packing_format": "auto_round:auto_gptq", | |
| "extra_config": { | |
| "model.language_model.layers.0.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.1.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.2.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.3.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.4.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.5.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.6.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.7.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.8.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.9.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.10.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.11.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.12.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.13.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.14.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.15.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.16.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.17.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.18.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.19.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.20.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.21.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.22.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.23.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.24.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.25.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.26.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.27.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.28.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.29.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.30.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.31.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.32.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.33.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.34.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.35.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.36.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.37.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.38.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.39.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.40.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.41.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.42.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.43.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.44.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.45.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.46.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.47.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.48.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.49.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.50.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.51.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.0.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.1.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.2.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.3.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.4.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.5.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.6.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.7.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.8.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.9.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.10.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.11.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.12.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.13.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.14.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.15.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.16.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.17.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.18.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.19.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.20.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.21.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.22.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.23.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.24.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.25.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.26.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.27.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.28.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.29.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.30.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.31.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.32.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.33.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.34.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.35.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.36.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.37.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.38.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.39.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.40.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.41.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.42.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.43.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.44.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.45.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.46.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.47.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.48.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.49.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.50.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.51.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| ".*embed_tokens.*": { | |
| "bits": 16 | |
| }, | |
| ".*vision_tower.*": { | |
| "bits": 16 | |
| }, | |
| ".*vision_adapter.*": { | |
| "bits": 16 | |
| }, | |
| ".*vision_projection.*": { | |
| "bits": 16 | |
| }, | |
| ".*self_attn\\.k_proj.*": { | |
| "bits": 16 | |
| }, | |
| ".*self_attn\\.v_proj.*": { | |
| "bits": 16 | |
| }, | |
| ".*self_attn\\.q_proj.*": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.0.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.1.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.2.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.3.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.4.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.5.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.6.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.7.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.8.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.9.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.10.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.11.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.12.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.13.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.14.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.15.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.16.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.17.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.18.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.19.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.20.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.21.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.22.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.23.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.24.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.25.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.26.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.27.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.28.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.29.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.30.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.31.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.32.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.33.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.34.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.35.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.36.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.37.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.38.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.39.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.40.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.41.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.42.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.43.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.44.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.45.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.46.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.47.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.48.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.49.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.50.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.language_model.layers.51.self_attn.q_proj": { | |
| "bits": 16 | |
| } | |
| } | |
| } |