Instructions to use MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2 with Transformers:

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

pipe = pipeline("text-generation", model="MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2")
model = AutoModelForCausalLM.from_pretrained("MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2")
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

vLLM

How to use MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2

SGLang

How to use MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2 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 "MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2" \
    --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": "MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2",
		"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 "MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2" \
        --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": "MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2 with Docker Model Runner:
```
docker model run hf.co/MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2
```

Configuration Parsing Warning:In config.json: "quantization_config.bits" must be an integer

Quantization

Quantized using the default exllamav2 (0.2.9) quantization process.
Original model: https://huggingface.co/Downtown-Case/Star-Command-R-Lite-32B-v1
exllamav2: https://github.com/turboderp-org/exllamav2

Original model card of Star-Command-R-Lite-32B-v1

A simple SLERP merge of TheDrummer's Star-Command with its base model, to tone it down and "keep" more of Command-R.

https://huggingface.co/TheDrummer/Star-Command-R-32B-v1

Created using mergekit.

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

TheDrummer/Star-Command-R-32B-v1
CohereForAI/c4ai-command-r-08-2024

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: TheDrummer/Star-Command-R-32B-v1
  - model: CohereForAI/c4ai-command-r-08-2024
merge_method: slerp
parameters:
  t:
    - value: 0.5
base_model: CohereForAI/c4ai-command-r-08-2024
dtype: bfloat16

Downloads last month: 1

Model tree for MetaphoricalCode/Star-Command-R-Lite-32B-v1-4.25bpw-h8-exl2

Base model

Downtown-Case/Star-Command-R-Lite-32B-v1

Quantized

(8)

this model