Instructions to use sankalphs/noir-verdict-nemotron-4b-merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sankalphs/noir-verdict-nemotron-4b-merged with Transformers:

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

pipe = pipeline("text-generation", model="sankalphs/noir-verdict-nemotron-4b-merged", 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("sankalphs/noir-verdict-nemotron-4b-merged", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("sankalphs/noir-verdict-nemotron-4b-merged", trust_remote_code=True)
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 sankalphs/noir-verdict-nemotron-4b-merged with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/sankalphs/noir-verdict-nemotron-4b-merged

SGLang

How to use sankalphs/noir-verdict-nemotron-4b-merged 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 "sankalphs/noir-verdict-nemotron-4b-merged" \
    --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": "sankalphs/noir-verdict-nemotron-4b-merged",
		"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 "sankalphs/noir-verdict-nemotron-4b-merged" \
        --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": "sankalphs/noir-verdict-nemotron-4b-merged",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use sankalphs/noir-verdict-nemotron-4b-merged with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for sankalphs/noir-verdict-nemotron-4b-merged to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for sankalphs/noir-verdict-nemotron-4b-merged to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for sankalphs/noir-verdict-nemotron-4b-merged to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="sankalphs/noir-verdict-nemotron-4b-merged",
    max_seq_length=2048,
)

Docker Model Runner
How to use sankalphs/noir-verdict-nemotron-4b-merged with Docker Model Runner:
```
docker model run hf.co/sankalphs/noir-verdict-nemotron-4b-merged
```

noir-verdict-nemotron-4b-merged

How to use

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "sankalphs/noir-verdict-nemotron-4b-merged"
tok = AutoTokenizer.from_pretrained(repo, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo, torch_dtype=torch.bfloat16, trust_remote_code=True,
).cuda().eval()

Chat template

The chat template is the Nemotron 3 chat template, with enable_thinking=False baked in. The system prompt for an active interrogation is built by engine/prompts.py:build_system_prompt(...).

messages = [
    {"role": "system", "content": "You are Greta Lindholm, junior continuity writer at WJBK. ..."},
    {"role": "user",   "content": "Where were you at the time of the theft?"},
]
text = tok.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=False)

Inference tips

n_ctx ≥ 4096
temperature 0.6–0.7, top_p 0.9–0.95
max_new_tokens 180–280 per turn
Stop on <|im_end|>

How it was trained

Image: nvidia/cuda:12.8.1-devel-ubuntu22.04 + Python 3.13
Pip deps: torch>=2.8.0, triton>=3.4.0, unsloth_zoo[base] @ git+https://github.com/unslothai/unsloth-zoo, unsloth[base] @ git+https://github.com/unslothai/unsloth, --torch-backend=cu128
Native: causal-conv1d==1.6.2.post1 and mamba-ssm==2.3.2.post1 compiled from source with --no-build-isolation, CC=gcc, CXX=g++ (no prebuilt cu128 + Py3.13 wheel exists)
Trainer: TRL SFTTrainer, packing, bf16, Unsloth LoRA (r=16, alpha=32, lr=2e-4 cosine, bs=2 grad_accum=8, 240 steps)
Orchestrator: train/modal_finetune.py

5-case smoke results (A10G, `--n-gpu-layers 99`)

case	suspect	personality	truth_mode	failure_flags
0	Greta Lindholm	nervous	lie	none
37	(37, 1)	helpful	partial_truth	none
113	(113, 2)	arrogant	truth	none
241	(241, 3)	evasive	deflect	none
497	Greta Lindholm	nervous	lie	none

Pace: ~125 tokens/sec on A10G. No role-token leaks, no leaked <think> blocks, no overlong generations.

Companion artifacts

LoRA: sankalphs/noir-verdict-nemotron-4b-lora
Merged BF16: sankalphs/noir-verdict-nemotron-4b-merged (7.95 GB)
Q4_K_M GGUF: sankalphs/noir-verdict-nemotron-4b-gguf (2.84 GB)
Traces: sankalphs/noir-verdict-traces
App: build-small-hackathon/noir-verdict

License

Apache-2.0. The base Nemotron 3 Nano weights are governed by NVIDIA's model license; the adapter and training code in this repo are Apache-2.0.

Downloads last month: 18

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for sankalphs/noir-verdict-nemotron-4b-merged

Base model

nvidia/NVIDIA-Nemotron-Nano-12B-v2-Base

Finetuned

nvidia/NVIDIA-Nemotron-Nano-12B-v2

Finetuned

nvidia/NVIDIA-Nemotron-Nano-9B-v2

Finetuned

nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16

Finetuned

unsloth/NVIDIA-Nemotron-3-Nano-4B

Finetuned

(3)

this model

sankalphs
/

noir-verdict-nemotron-4b-merged

noir-verdict-nemotron-4b-merged

How to use

Chat template

Inference tips

How it was trained

5-case smoke results (A10G, `--n-gpu-layers 99`)

Companion artifacts

License

Model tree for sankalphs/noir-verdict-nemotron-4b-merged

Spaces using sankalphs/noir-verdict-nemotron-4b-merged 2

noir-verdict-nemotron-4b-merged

How to use

Chat template

Inference tips

How it was trained

5-case smoke results (A10G, --n-gpu-layers 99)

Companion artifacts

License

Model tree for sankalphs/noir-verdict-nemotron-4b-merged

Spaces using sankalphs/noir-verdict-nemotron-4b-merged 2

5-case smoke results (A10G, `--n-gpu-layers 99`)