Instructions to use xriminact/llama-3-8b-instruct-openvino-int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xriminact/llama-3-8b-instruct-openvino-int4 with Transformers:

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

pipe = pipeline("text-generation", model="xriminact/llama-3-8b-instruct-openvino-int4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

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

tokenizer = AutoTokenizer.from_pretrained("xriminact/llama-3-8b-instruct-openvino-int4")
model = AutoModelForCausalLM.from_pretrained("xriminact/llama-3-8b-instruct-openvino-int4", 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 xriminact/llama-3-8b-instruct-openvino-int4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "xriminact/llama-3-8b-instruct-openvino-int4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "xriminact/llama-3-8b-instruct-openvino-int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/xriminact/llama-3-8b-instruct-openvino-int4

SGLang

How to use xriminact/llama-3-8b-instruct-openvino-int4 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 "xriminact/llama-3-8b-instruct-openvino-int4" \
    --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": "xriminact/llama-3-8b-instruct-openvino-int4",
		"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 "xriminact/llama-3-8b-instruct-openvino-int4" \
        --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": "xriminact/llama-3-8b-instruct-openvino-int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use xriminact/llama-3-8b-instruct-openvino-int4 with Docker Model Runner:
```
docker model run hf.co/xriminact/llama-3-8b-instruct-openvino-int4
```

Usage

from transformers import AutoConfig, AutoTokenizer
from optimum.intel.openvino import OVModelForCausalLM

ov_config = {"PERFORMANCE_HINT": "LATENCY", "NUM_STREAMS": "1", "CACHE_DIR": "", "INFERENCE_PRECISION_HINT": "f16"}

tok = AutoTokenizer.from_pretrained("xriminact/llama-3-8b-instruct-openvino-int4", trust_remote_code=True)

ov_model = OVModelForCausalLM.from_pretrained(
    "xriminact/llama-3-8b-instruct-openvino-int4",
    device="GPU",
    ov_config=ov_config,
    config=AutoConfig.from_pretrained("xriminact/llama-3-8b-instruct-openvino-int4", trust_remote_code=True),
    trust_remote_code=True,
)

test_string = "What is OpenVino?"
input_tokens = tok(test_string, return_tensors="pt")
answer = ov_model.generate(**input_tokens, max_new_tokens=200)
print(tok.batch_decode(answer, skip_special_tokens=True)[0])

Downloads last month: 12