Instructions to use ishorn5/RTLCoder-v1.1-gptq-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ishorn5/RTLCoder-v1.1-gptq-4bit with Transformers:

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

pipe = pipeline("text-generation", model="ishorn5/RTLCoder-v1.1-gptq-4bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ishorn5/RTLCoder-v1.1-gptq-4bit")
model = AutoModelForCausalLM.from_pretrained("ishorn5/RTLCoder-v1.1-gptq-4bit")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ishorn5/RTLCoder-v1.1-gptq-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ishorn5/RTLCoder-v1.1-gptq-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ishorn5/RTLCoder-v1.1-gptq-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ishorn5/RTLCoder-v1.1-gptq-4bit

SGLang

How to use ishorn5/RTLCoder-v1.1-gptq-4bit 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 "ishorn5/RTLCoder-v1.1-gptq-4bit" \
    --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": "ishorn5/RTLCoder-v1.1-gptq-4bit",
		"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 "ishorn5/RTLCoder-v1.1-gptq-4bit" \
        --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": "ishorn5/RTLCoder-v1.1-gptq-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ishorn5/RTLCoder-v1.1-gptq-4bit with Docker Model Runner:
```
docker model run hf.co/ishorn5/RTLCoder-v1.1-gptq-4bit
```

TypeError: BaseGPTQForCausalLM.generate() takes 1 positional argument but 2 were given

by XieTangent - opened Jan 20, 2024

Discussion

XieTangent

Jan 20, 2024

When I try to test the RTLCoder-gptq-4bit
Python tell me that input_ids regarded as two arguments
but isn't following code the solution to this problem?

input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to(0)

Or am I mistaken?
how can i solve this problem?

ishorn5

Owner Jan 21, 2024

•

edited Jan 21, 2024

Hi, thanks for your issue! The following code should work,
'''
from transformers import AutoTokenizer
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
prompt = "Please act as a professional verilog designer and provide a half adder. \nmodule half_adder\n(input a, \ninput b, \noutput sum, \noutput carry);\n"

tokenizer = AutoTokenizer.from_pretrained("ishorn5/RTLCoder-v1.1-gptq-4bit")
model = AutoGPTQForCausalLM.from_quantized("ishorn5/RTLCoder-v1.1-gptq-4bit", device=0)
model.eval()
inputs = tokenizer(prompt, return_tensors="pt").to(0)
sample = model.generate(**inputs, max_length=512, temperature=0.5, top_p=0.9, pad_token_id=model.config.eos_token_id,)
print(tokenizer.decode(sample[0], skip_special_tokens=True))
'''

XieTangent

Jan 22, 2024

Appreciate the clarification

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment