NVIDIA Nemotron 3.5 Lightning 30B-A3B — oQ8

A native Apple-silicon conversion of nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16, quantized with oMLX's oQ8 affine recipe and packaged for MLX-LM and oMLX.

Original model · NVIDIA Nemotron · MLX-LM · OpenMDW 1.1 license

About this conversion

This repository contains an oMLX oQ8 affine conversion of NVIDIA Nemotron 3.5 Lightning. The source is a 30B-total / 3B-active hybrid mixture-of-experts model that interleaves Mamba-2, sparse MoE, and attention layers. The upstream tokenizer, chat template, and generation configuration are preserved.

[!NOTE] oQ8 uses uniform 8-bit affine weights with a group size of 64.

Apple-silicon performance

This checkpoint was load-tested and generation-tested on the following machine:

A warmed local test produced:

The decode figure is the median of three greedy 256-token runs after a 32-token Metal-kernel warm-up. It is a practical local reference, not a controlled cross-platform benchmark. Prompt length, context growth, sampler settings, memory pressure, thermal state, and MLX/oMLX versions can materially change performance.

Quick start with MLX-LM

Install recent MLX-LM and Hugging Face tooling:

python -m pip install -U mlx-lm huggingface_hub

Run directly from the Hub:

mlx_lm.generate \
  --model TensorFold/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8 \
  --prompt "Explain why hybrid Mamba and MoE architectures are efficient." \
  --max-tokens 512 \
  --temp 1.0 \
  --top-p 0.95

Reasoning mode is enabled by the upstream chat template by default. To disable it:

mlx_lm.generate \
  --model TensorFold/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8 \
  --chat-template-config '{"enable_thinking": false}' \
  --prompt "Write a short hello-world program in Swift." \
  --max-tokens 256

Python usage:

from mlx_lm import load, generate


model, tokenizer = load("TensorFold/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8")
messages = [
    {"role": "user", "content": "Explain sparse mixture-of-experts routing."}
]
prompt = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    tokenize=False,
    enable_thinking=False,
)
print(generate(model, tokenizer, prompt=prompt, max_tokens=512))

To download the repository first:

hf download TensorFold/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8 \
  --local-dir ~/.omlx/models/TensorFold/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8

Using it with oMLX

1. Place the downloaded model at ~/.omlx/models/TensorFold/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8.
2. Refresh the oMLX model registry.
3. Load NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8 and use the normal chat UI or OpenAI-compatible endpoint.

Example request:

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OMLX_API_KEY" \
  -d '{
    "model": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8",
    "messages": [{"role": "user", "content": "Say hello from Nemotron on MLX."}],
    "temperature": 1.0,
    "top_p": 0.95,
    "max_tokens": 128
  }'

For long prompts, begin with a conservative context limit and increase it while watching memory pressure. The configured 256K context is a model capability, not a guarantee that every host can prefill that context within its available unified memory.

Architecture

Nemotron 3.5 Lightning is a hybrid sparse model designed for efficient agentic and reasoning workloads.

The upstream release is intended for coding, tool use, reasoning, research, and customization. For NVIDIA's evaluations, deployment guidance, intended use, limitations, safety information, and full architecture discussion, see the original model card.

Conversion and validation notes

• Source weights: NVIDIA's BF16 checkpoint.
• Quantization group size: 64.
• Quantization mode: affine.
• The upstream chat_template.jinja is preserved.
• All 729 converted tensors and every indexed shard were checked locally.
• The model was loaded and exercised through end-to-end generation on Apple silicon.
• Quantization can reduce output quality relative to BF16; use a higher-precision variant when quality matters more than memory use.

This is a community conversion, not an official NVIDIA release. Validate quality and numerical behavior on your own representative workload before production use.

License and attribution

The upstream model is released under the OpenMDW License Agreement, version 1.1. A copy is included in this repository; review it before use or redistribution.

All model design, training, benchmark, and upstream documentation credit belongs to NVIDIA and the original contributors. The MLX conversion, Apple-silicon validation, and packaging are provided by TensorFold.

Choose for your Mac

64GB Macs · 128GB Macs · 256GB Macs

Published peak memory: 33.73 GB; estimated starting tier: 64GB, leaving about 30 GB nominal headroom. The collections use published M3 Studio peaks with at least 25% nominal headroom; fit on other Macs is an estimate, and full context is not guaranteed. Start with short context and one request.

Runtime and evidence

The exact tested oMLX application version is not recorded here; a library version is not an app version. The original performance tables retain their benchmark conditions and speed figures; this documentation update adds no new test results.

Quick start and demo prompt

hf download TensorFold/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8 --local-dir ./models/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-oQ8

Add the downloaded folder to oMLX model directories, refresh the list, and follow this card's architecture and MTP compatibility requirements before loading.

Try this in a new chat with a 128-token output limit:

Explain why the sky looks blue in three short sentences.

This is a demo prompt to try, not a recorded successful run; a captured demonstration for this documentation update is not yet available.

Follow TensorFold for new Apple Silicon releases and fixes.