Instructions to use vcruz305/DeepNemotron-3.5-Lightning-MXFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use vcruz305/DeepNemotron-3.5-Lightning-MXFP4 with Transformers:

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

pipe = pipeline("text-generation", model="vcruz305/DeepNemotron-3.5-Lightning-MXFP4")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("vcruz305/DeepNemotron-3.5-Lightning-MXFP4", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use vcruz305/DeepNemotron-3.5-Lightning-MXFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "vcruz305/DeepNemotron-3.5-Lightning-MXFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "vcruz305/DeepNemotron-3.5-Lightning-MXFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/vcruz305/DeepNemotron-3.5-Lightning-MXFP4

SGLang

How to use vcruz305/DeepNemotron-3.5-Lightning-MXFP4 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 "vcruz305/DeepNemotron-3.5-Lightning-MXFP4" \
    --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": "vcruz305/DeepNemotron-3.5-Lightning-MXFP4",
		"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 "vcruz305/DeepNemotron-3.5-Lightning-MXFP4" \
        --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": "vcruz305/DeepNemotron-3.5-Lightning-MXFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use vcruz305/DeepNemotron-3.5-Lightning-MXFP4 with Docker Model Runner:
```
docker model run hf.co/vcruz305/DeepNemotron-3.5-Lightning-MXFP4
```

DeepNemotron-3.5-Lightning-MXFP4

MXFP4 of the merged M5 Lightning checkpoint.

Adapter: vcruz305/DeepNemotron-3.5-Lightning
BF16: vcruz305/DeepNemotron-3.5-Lightning-BF16
K-quants: vcruz305/DeepNemotron-3.5-Lightning-GGUF
NVFP4: vcruz305/DeepNemotron-3.5-Lightning-NVFP4

Train set

Same SFT as the adapter: r0b0tlab/deepseek-v4-pro-0813-agentic config sft_openai, train split 19,072 rows, max seq 512.

DS4 is a verifiable-first agentic corpus from DeepSeek-V4-Pro 0813 (official API, thinking on). Rows pass a deterministic verifier before admission. 13 families cover tool-schema adherence with sandbox execution, constraint following, structured outputs, multi-turn state, math/science recomputation, multilingual turns, planning, long-context citation, delegation, memory compression, and hidden-test code. Published split 19,072 / 1,070 / 1,065.

Train (MEASURED)

Modal A100-80, Unsloth 2026.8.18, torch 2.7.1+cu118
Q4′ LoRA r=16 α=32 · attn q/k/v/o + Mamba in_proj + shared-expert Linears
19,072 / 19,072 steps, epoch 1.0, train_loss 0.151
MXFP4 is a post-merge quant of that adapter

License

NVIDIA Nemotron 3.5 Lightning (OpenMDW) and dataset license other (DeepSeek API-adjacent). Use is under the base card and the dataset card.

Downloads last month: -; Downloads are not tracked for this model. How to track

Model tree for vcruz305/DeepNemotron-3.5-Lightning-MXFP4

Base model

unsloth/NVIDIA-Nemotron-3.5-Lightning-30B-A3B

Finetuned

vcruz305/DeepNemotron-3.5-Lightning-BF16

Quantized

(1)

this model

vcruz305
/

DeepNemotron-3.5-Lightning-MXFP4

DeepNemotron-3.5-Lightning-MXFP4

Train set

Train (MEASURED)

License

Model tree for vcruz305/DeepNemotron-3.5-Lightning-MXFP4

Dataset used to train vcruz305/DeepNemotron-3.5-Lightning-MXFP4