Instructions to use mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128 with Transformers:

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

pipe = pipeline("text-generation", model="mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128")
model = AutoModel.from_pretrained("mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128

SGLang

How to use mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128 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 "mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128" \
    --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": "mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128" \
        --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": "mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128 with Docker Model Runner:
```
docker model run hf.co/mgaruccio/Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128
```

Muse-Glimmer-30B-assistant-GPTQ-Int4-sym-G128

File size: 2,331 Bytes

8a61e51

{
  "architectures": [
    "MuseGlimmerAssistantModel"
  ],
  "attention_dropout": 0,
  "block_size": 16,
  "bos_token_id": 200000,
  "dtype": "bfloat16",
  "eos_token_id": 200001,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 6656,
  "intermediate_size": 19968,
  "layer_types": [
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention"
  ],
  "mask_token_id": 201818,
  "max_position_embeddings": 131072,
  "model_type": "muse_glimmer_assistant",
  "num_attention_heads": 32,
  "num_hidden_layers": 5,
  "num_key_value_heads": 8,
  "pad_token_id": 200018,
  "quantization_config": {
    "bits": 4,
    "checkpoint_format": "gptq",
    "desc_act": false,
    "format": "gptq",
    "group_size": 128,
    "lm_head": false,
    "meta": {
      "act_group_aware": true,
      "auto_forward_data_parallel": true,
      "damp_auto_increment": 0.01,
      "damp_percent": 0.05,
      "dense_vram_strategy": "exclusive",
      "dense_vram_strategy_devices": null,
      "fallback": {
        "smooth": null,
        "strategy": "rtn",
        "threshold": "0.5%"
      },
      "foem": null,
      "gc_mode": "interval",
      "gptaq": null,
      "hessian": {
        "chunk_bytes": null,
        "chunk_size": null,
        "staging_dtype": "float32"
      },
      "mock_quantization": false,
      "moe_vram_strategy": "exclusive",
      "moe_vram_strategy_devices": null,
      "mse": 0.0,
      "offload_to_disk": true,
      "offload_to_disk_path": "/tmp/muse-dflash-assistant-gptq-offload",
      "pack_impl": "cpu",
      "quantizer": [
        "gptqmodel:7.3.2"
      ],
      "static_groups": false,
      "true_sequential": true,
      "uri": "https://github.com/modelcloud/gptqmodel",
      "wait_for_submodule_finalizers": false
    },
    "method": "gptq",
    "modules_in_block_to_quantize": [
      "self_attn.qkv_proj",
      "self_attn.o_proj",
      "mlp.gate_up_proj",
      "mlp.down_proj"
    ],
    "pack_dtype": "int32",
    "quant_method": "gptq",
    "sym": true
  },
  "rms_norm_eps": 1e-05,
  "rope_parameters": {
    "rope_theta": 500000.0,
    "rope_type": "default"
  },
  "sliding_window": 2048,
  "target_layer_ids": [
    1,
    13,
    25,
    37,
    49
  ],
  "transformers_version": "5.15.1",
  "use_cache": false
}