Instructions to use rodrigoramosrs/veriloop-coder-e2-nvfp4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rodrigoramosrs/veriloop-coder-e2-nvfp4 with Transformers:

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

pipe = pipeline("text-generation", model="rodrigoramosrs/veriloop-coder-e2-nvfp4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("rodrigoramosrs/veriloop-coder-e2-nvfp4")
model = AutoModelForCausalLM.from_pretrained("rodrigoramosrs/veriloop-coder-e2-nvfp4", 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 rodrigoramosrs/veriloop-coder-e2-nvfp4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rodrigoramosrs/veriloop-coder-e2-nvfp4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rodrigoramosrs/veriloop-coder-e2-nvfp4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rodrigoramosrs/veriloop-coder-e2-nvfp4

SGLang

How to use rodrigoramosrs/veriloop-coder-e2-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 "rodrigoramosrs/veriloop-coder-e2-nvfp4" \
    --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": "rodrigoramosrs/veriloop-coder-e2-nvfp4",
		"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 "rodrigoramosrs/veriloop-coder-e2-nvfp4" \
        --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": "rodrigoramosrs/veriloop-coder-e2-nvfp4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use rodrigoramosrs/veriloop-coder-e2-nvfp4 with Docker Model Runner:
```
docker model run hf.co/rodrigoramosrs/veriloop-coder-e2-nvfp4
```

veriloop-coder-e2-nvfp4 / scripts /validate_nvfp4.py

rodrigoramosrs

Upload scripts/validate_nvfp4.py with huggingface_hub

76376a0 verified 1 day ago

Raw

History Blame Contribute Delete

1.72 kB

	"""Structural validation for a modelopt NVFP4 checkpoint.

	Checks that every tensor in model.safetensors.index.json exists in the
	shards (and vice versa) and prints the dtype layout + quant config.

	Usage:
	python validate_nvfp4.py [MODEL_DIR] (default: current directory)
	"""
	import collections
	import json
	import os
	import struct
	import sys

	DD = sys.argv[1] if len(sys.argv) > 1 else "."
	idx = json.load(open(os.path.join(DD, "model.safetensors.index.json")))
	wm = idx["weight_map"]

	# headers of each shard (no tensor data is read)
	have = {}
	for fname in set(wm.values()):
	with open(os.path.join(DD, fname), "rb") as f:
	n = struct.unpack("<Q", f.read(8))[0]
	header = json.loads(f.read(n))
	for t in header:
	if t != "__metadata__":
	have[t] = (fname, header[t]["dtype"], header[t]["shape"])

	missing = [t for t in wm if t not in have]
	extra = [t for t in have if t not in wm]
	print("tensors in index:", len(wm))
	print("tensors in shards:", len(have))
	print("missing:", len(missing), missing[:5])
	print("extra:", len(extra), extra[:5])

	dtypes = collections.Counter(v[1] for v in have.values())
	print("dtypes:", dict(dtypes))
	# packed NVFP4 weights show up as U8 with per-block scale tensors
	suff = collections.Counter()
	for t in have:
	s = t.split(".")[-1]
	suff[s] += 1
	print("common suffixes:", suff.most_common(12))
	q = json.load(open(os.path.join(DD, "hf_quant_config.json")))
	print("quant_algo:", q["quantization"]["quant_algo"],
	"\| group_size:", q["quantization"]["group_size"])

	assert not missing and not extra, "shard/index mismatch!"
	assert q["quantization"]["quant_algo"] == "NVFP4", "not an NVFP4 checkpoint!"
	print("OK - valid NVFP4 checkpoint")