Instructions to use abdelkader-dev/Quark-flash-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use abdelkader-dev/Quark-flash-1 with Transformers:

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

pipe = pipeline("text-generation", model="abdelkader-dev/Quark-flash-1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("abdelkader-dev/Quark-flash-1")
model = AutoModelForCausalLM.from_pretrained("abdelkader-dev/Quark-flash-1", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use abdelkader-dev/Quark-flash-1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "abdelkader-dev/Quark-flash-1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "abdelkader-dev/Quark-flash-1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/abdelkader-dev/Quark-flash-1

SGLang

How to use abdelkader-dev/Quark-flash-1 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 "abdelkader-dev/Quark-flash-1" \
    --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": "abdelkader-dev/Quark-flash-1",
		"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 "abdelkader-dev/Quark-flash-1" \
        --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": "abdelkader-dev/Quark-flash-1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use abdelkader-dev/Quark-flash-1 with Docker Model Runner:
```
docker model run hf.co/abdelkader-dev/Quark-flash-1
```

Quark-flash-1 / README.md

abdelkader-dev

Update README.md

d853e1a verified 8 days ago

preview code

Raw

History Blame Contribute Delete

833 Bytes

metadata

license: mit
language:
  - en
pipeline_tag: text-generation
library_name: transformers
base_model:
  - abdelkader-dev/Quark-pre-30M
tags:
  - Quark
  - Quark-flash
  - mini-model
  - nano-model

Quark-flash-1 is a small language model (SLM) designed for quick responses and lightweight performance on most low-spec devices.

The model covers questions about medical emergencies and some survival techniques in emergency situations, accidents, and injuries.

Questions should be short and clear for the model to provide an accurate answer, quickly indicating how to act.

Because the model is small, its answers may not be comprehensive or precise.

The model achieved a score of 9.5/10 in internal tests of several questions covering (accident response, fires, survival in harsh environments such as forests, and serious injuries).