Instructions to use Gaivoronsky/ruGPT-3.5-13B-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Gaivoronsky/ruGPT-3.5-13B-8bit with Transformers:

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

pipe = pipeline("text-generation", model="Gaivoronsky/ruGPT-3.5-13B-8bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Gaivoronsky/ruGPT-3.5-13B-8bit")
model = AutoModelForCausalLM.from_pretrained("Gaivoronsky/ruGPT-3.5-13B-8bit")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Gaivoronsky/ruGPT-3.5-13B-8bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Gaivoronsky/ruGPT-3.5-13B-8bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gaivoronsky/ruGPT-3.5-13B-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Gaivoronsky/ruGPT-3.5-13B-8bit

SGLang

How to use Gaivoronsky/ruGPT-3.5-13B-8bit 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 "Gaivoronsky/ruGPT-3.5-13B-8bit" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gaivoronsky/ruGPT-3.5-13B-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Gaivoronsky/ruGPT-3.5-13B-8bit" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gaivoronsky/ruGPT-3.5-13B-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Gaivoronsky/ruGPT-3.5-13B-8bit with Docker Model Runner:
```
docker model run hf.co/Gaivoronsky/ruGPT-3.5-13B-8bit
```

This is a generative model converted to fp16 format based on ai-forever/ruGPT-3.5-13B

Examples of usage

from transformers import AutoTokenizer
from auto_gptq import AutoGPTQForCausalLM

model = AutoGPTQForCausalLM.from_quantized('Gaivoronsky/ruGPT-3.5-13B-8bit', device="cuda:0", use_triton=False)
tokenizer = AutoTokenizer.from_pretrained('Gaivoronsky/ruGPT-3.5-13B-8bit')

request = "Человек: Сколько весит жираф? Помощник: "
encoded_input = tokenizer(request, return_tensors='pt', \
                          add_special_tokens=False).to('cuda')
output = model.generate(
    **encoded_input,
    num_beams=2,
    do_sample=True,
    max_new_tokens=100
)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Downloads last month: 8