Instructions to use nguyennm1024/dana-nemo-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nguyennm1024/dana-nemo-4b with Transformers:

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

pipe = pipeline("text-generation", model="nguyennm1024/dana-nemo-4b", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nguyennm1024/dana-nemo-4b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("nguyennm1024/dana-nemo-4b", trust_remote_code=True, 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nguyennm1024/dana-nemo-4b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nguyennm1024/dana-nemo-4b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nguyennm1024/dana-nemo-4b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nguyennm1024/dana-nemo-4b

SGLang

How to use nguyennm1024/dana-nemo-4b 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 "nguyennm1024/dana-nemo-4b" \
    --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": "nguyennm1024/dana-nemo-4b",
		"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 "nguyennm1024/dana-nemo-4b" \
        --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": "nguyennm1024/dana-nemo-4b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nguyennm1024/dana-nemo-4b with Docker Model Runner:
```
docker model run hf.co/nguyennm1024/dana-nemo-4b
```

dana-nemo-4b

A 4B-parameter causal language model based on the Nemotron-H architecture, released in FP8 quantized format.

Model Details

Architecture: NemotronHForCausalLM (hybrid Mamba/Attention)
Parameters: ~4B
Hidden size: 3136
Layers: 42
Attention heads: 40 (8 KV heads)
Context length: 262,144 tokens
Vocabulary size: 131,072
Precision: FP8 (weights and KV cache), bfloat16 compute dtype
License: NVIDIA Nemotron Open Model License

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "<org>/dana-nemo-4b"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True, device_map="auto")

messages = [{"role": "user", "content": "Hello!"}]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
output = model.generate(inputs, max_new_tokens=256)
print(tokenizer.decode(output[0], skip_special_tokens=True))

This model requires trust_remote_code=True due to the custom Nemotron-H modeling code included in this repository.

License

This model is released under the NVIDIA Nemotron Open Model License.

Downloads last month: 410

Safetensors

Model size

4B params

Tensor type

F32

BF16

F8_E4M3

Model tree for nguyennm1024/dana-nemo-4b

Unable to build the model tree, the base model loops to the model itself. Learn more.