Instructions to use Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw with Transformers:

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

pipe = pipeline("text-generation", model="Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw")
model = AutoModelForCausalLM.from_pretrained("Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw")
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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw

SGLang

How to use Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw 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 "Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw" \
    --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": "Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw",
		"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 "Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw" \
        --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": "Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw with Docker Model Runner:
```
docker model run hf.co/Downtown-Case/Star-Command-R-Lite-32B-v1-exl2-4bpw
```

Star-Command-R-Lite-32B-v1-exl2-4bpw

File size: 1,042 Bytes

142362d

{
    "_name_or_path": "/home/alpha/Models/Raw/CohereForAI_c4ai-command-r-08-2024",
    "architectures": [
        "CohereForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 5,
    "eos_token_id": 255001,
    "hidden_act": "silu",
    "hidden_size": 8192,
    "initializer_range": 0.02,
    "intermediate_size": 24576,
    "layer_norm_eps": 1e-05,
    "logit_scale": 0.0625,
    "max_position_embeddings": 131072,
    "model_type": "cohere",
    "num_attention_heads": 64,
    "num_hidden_layers": 40,
    "num_key_value_heads": 8,
    "pad_token_id": 0,
    "rope_theta": 4000000,
    "torch_dtype": "bfloat16",
    "transformers_version": "4.45.0.dev0",
    "use_cache": true,
    "use_qk_norm": false,
    "vocab_size": 256000,
    "quantization_config": {
        "quant_method": "exl2",
        "version": "0.2.0",
        "bits": 4.0,
        "head_bits": 6,
        "calibration": {
            "rows": 115,
            "length": 2048,
            "dataset": "(default)"
        }
    }
}