Instructions to use IlyasMoutawwakil/tiny-random-Llama-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IlyasMoutawwakil/tiny-random-Llama-NVFP4 with Transformers:

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

pipe = pipeline("text-generation", model="IlyasMoutawwakil/tiny-random-Llama-NVFP4")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("IlyasMoutawwakil/tiny-random-Llama-NVFP4")
model = AutoModelForCausalLM.from_pretrained("IlyasMoutawwakil/tiny-random-Llama-NVFP4", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use IlyasMoutawwakil/tiny-random-Llama-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IlyasMoutawwakil/tiny-random-Llama-NVFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IlyasMoutawwakil/tiny-random-Llama-NVFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/IlyasMoutawwakil/tiny-random-Llama-NVFP4

SGLang

How to use IlyasMoutawwakil/tiny-random-Llama-NVFP4 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 "IlyasMoutawwakil/tiny-random-Llama-NVFP4" \
    --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": "IlyasMoutawwakil/tiny-random-Llama-NVFP4",
		"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 "IlyasMoutawwakil/tiny-random-Llama-NVFP4" \
        --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": "IlyasMoutawwakil/tiny-random-Llama-NVFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use IlyasMoutawwakil/tiny-random-Llama-NVFP4 with Docker Model Runner:
```
docker model run hf.co/IlyasMoutawwakil/tiny-random-Llama-NVFP4
```

tiny-random-Llama-NVFP4

A tiny random model for testing, shrunk from nvidia/Llama-3.1-8B-Instruct-FP4: the same architecture, quantization config and checkpoint layout at test sizes. Its key patterns, dtypes and tensor ranks match the real checkpoint's (scripts/extract_layout.py).

modelopt NVFP4 on every decoder linear (packed U8, E4M3 weight_scale per 16, F32 weight_scale_2, calibrated F32 input_scale) plus the FP8 KV cache's k_scale / v_scale. Quantized with modelopt's NVFP4QTensor.

reference/ holds the same weights dequantized to bf16, under the unquantized model's keys: the reference to compare logits against, so a test measures what the load path and kernels add, not the quantization itself.

The weights are random; the outputs mean nothing. scripts/ rebuilds it from the real checkpoint's config.json.

Downloads last month: 607

Safetensors

Model size

1.12M params

Tensor type

BF16