Instructions to use torchao-dev/opt-125m-float8dq-row-0.13-dev with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use torchao-dev/opt-125m-float8dq-row-0.13-dev with Transformers:

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

pipe = pipeline("text-generation", model="torchao-dev/opt-125m-float8dq-row-0.13-dev", device_map="auto")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("torchao-dev/opt-125m-float8dq-row-0.13-dev")
model = AutoModelForCausalLM.from_pretrained("torchao-dev/opt-125m-float8dq-row-0.13-dev", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use torchao-dev/opt-125m-float8dq-row-0.13-dev with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "torchao-dev/opt-125m-float8dq-row-0.13-dev"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "torchao-dev/opt-125m-float8dq-row-0.13-dev",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/torchao-dev/opt-125m-float8dq-row-0.13-dev

SGLang

How to use torchao-dev/opt-125m-float8dq-row-0.13-dev 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 "torchao-dev/opt-125m-float8dq-row-0.13-dev" \
    --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": "torchao-dev/opt-125m-float8dq-row-0.13-dev",
		"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 "torchao-dev/opt-125m-float8dq-row-0.13-dev" \
        --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": "torchao-dev/opt-125m-float8dq-row-0.13-dev",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use torchao-dev/opt-125m-float8dq-row-0.13-dev with Docker Model Runner:
```
docker model run hf.co/torchao-dev/opt-125m-float8dq-row-0.13-dev
```

opt-125m-float8dq-row-0.13-dev / README.md

TorchAO Testing

Update README.md (#1)

3a14d24 verified about 1 year ago

preview code

Raw

History Blame Contribute Delete

1.26 kB

	---
	library_name: transformers
	tags: []
	---

	```
	import torch
	from transformers import AutoModelForCausalLM, AutoTokenizer, TorchAoConfig

	model_id = "facebook/opt-125m"

	from torchao.quantization import Float8DynamicActivationFloat8WeightConfig, PerRow
	quant_config = Float8DynamicActivationFloat8WeightConfig(granularity=PerRow())

	quantization_config = TorchAoConfig(quant_type=quant_config)
	quantized_model = AutoModelForCausalLM.from_pretrained(
	model_id,
	device_map="cuda",
	torch_dtype=torch.bfloat16,
	quantization_config=quantization_config,
	)
	tokenizer = AutoTokenizer.from_pretrained(model_id)

	# Push to hub
	USER_ID = "torchao-testing"
	MODEL_NAME = model_id.split("/")[-1]
	save_to = f"{USER_ID}/{MODEL_NAME}-float8dq-row-0.13-dev"
	quantized_model.push_to_hub(save_to, safe_serialization=False)
	tokenizer.push_to_hub(save_to)

	# Manual Testing
	prompt = "Hey, are you conscious? Can you talk to me?"
	print("Prompt:", prompt)
	inputs = tokenizer(
	prompt,
	return_tensors="pt",
	).to("cuda")
	generated_ids = quantized_model.generate(**inputs, max_new_tokens=128)
	output_text = tokenizer.batch_decode(
	generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
	)
	print("Response:", output_text[0][len(prompt) :])

	```