Instructions to use emplitude/rubycon with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use emplitude/rubycon with Transformers:

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

pipe = pipeline("text-generation", model="emplitude/rubycon")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("emplitude/rubycon")
model = AutoModelForCausalLM.from_pretrained("emplitude/rubycon")
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

vLLM

How to use emplitude/rubycon with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/emplitude/rubycon

SGLang

How to use emplitude/rubycon 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 "emplitude/rubycon" \
    --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": "emplitude/rubycon",
		"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 "emplitude/rubycon" \
        --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": "emplitude/rubycon",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use emplitude/rubycon with Docker Model Runner:
```
docker model run hf.co/emplitude/rubycon
```

rubycon / modifyhash.py

emplitude

Upload model

d1a10bf verified almost 2 years ago

raw

history blame

1.01 kB

	from safetensors import safe_open
	from safetensors.torch import save_file
	import torch
	import hashlib

	filename = "model-00001-of-00001.safetensors"
	modified_model_name = "modified_model1.safetensors"


	with safe_open(filename, framework="pt") as f:
	tensors = {key: f.get_tensor(key) for key in f.keys()}

	def introduce_noise(tensor, noise_level=1e-8):
	noise = torch.randn(tensor.size()) * noise_level
	return (tensor + noise).to(tensor.dtype)

	for key in tensors:
	tensors[key] = introduce_noise(tensors[key])


	metadata = {
	"format": "pt" # Adjust based on actual format needed
	}

	save_file(tensors, modified_model_name, metadata=metadata)


	def compute_hash(filename):
	hasher = hashlib.sha256()
	with open(filename, "rb") as f:
	buf = f.read()
	hasher.update(buf)
	return hasher.hexdigest()

	original_hash = compute_hash(filename)
	modified_hash = compute_hash(modified_model_name)

	print(f"Original Hash: {original_hash}")
	print(f"Modified Hash: {modified_hash}")