Instructions to use OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

MLX LM

How to use OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M

Run Hermes

hermes

Atomic Chat

OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M

The standard profile. 8-bit attention/routers/shared experts, 4-bit routed experts.

JANG mixed-precision affine (JANG_4M, group size 64) conversion of nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 for Apple silicon (MLX).

NVIDIA Nemotron 3.5 Lightning is a 30B-total / 3B-active hybrid model: 52 layers of Mamba-2 + MoE + attention (23 Mamba, 23 MoE, 6 full-attention), 128 routed experts with 6 active plus 1 shared expert, and a 256K context window. Only 6 of 52 layers hold a KV cache, which is what makes long context affordable on a laptop.

Bundles in this series

Bundle	Avg bits	Size	Notes
`Nemotron-3.5-Lightning-30B-A3B-JANG_2L`	3.73	16.22 GiB	smallest
`Nemotron-3.5-Lightning-30B-A3B-JANG_4M`	4.12	17.72 GiB	recommended
`Nemotron-3.5-Lightning-30B-A3B-JANG_6M`	6.06	25.15 GiB	near-lossless
`Nemotron-3.5-Lightning-30B-A3B-MXFP8`	8.00	31.62 GiB	MX FP8 reference

Verified local row

The exact directory uploaded to this repository was loaded and generated on an M5 Max MacBook on 2026-08-11. Greedy decoding, prompts rendered through this bundle's own chat_template.jinja.

Model class: nemotron_h
Weight shards: 26
Weight bytes: 19024579336 (17.718 GiB)
Average bits/weight: 4.12
Load time: 1.9 seconds
Reasoning ON: 121.0 tok/s
Reasoning OFF: 47.5 tok/s
Stop behaviour: normal stop on every turn (no length stops)
Visible output: coherent on every turn
Tool calling: emitted a well-formed <tool_call> block

Probes: a factual question with reasoning on, the same with reasoning off, and a single-function tool-calling request.

Sampling

NVIDIA's recommended settings, stamped into both generation_config.json and jang_config.json so servers and loaders agree:

Parameter	Value
`temperature`	1.0
`top_p`	0.95
`top_k`	0 (disabled)
`repetition_penalty`	1.0 (none)
`eos_token_id`	`[2, 11]`

top_k is unspecified upstream and is stamped as 0 so downstream servers do not apply their own default. Both EOS ids matter: 11 (<|im_end|>) is the chat terminator and 2 (</s>) is the legacy one. Do not prepend BOS — add_bos_token is false.

Reasoning

Reasoning is on by default, matching upstream. There are exactly two states — there are no low/medium/high effort tiers.

# Reasoning ON (default)
prompt = tokenizer.apply_chat_template(msgs, add_generation_prompt=True,
                                       tokenize=False, enable_thinking=True)
# Reasoning OFF
prompt = tokenizer.apply_chat_template(msgs, add_generation_prompt=True,
                                       tokenize=False, enable_thinking=False)

The chat template opens the reasoning rail itself: with reasoning on the prompt ends in <think>\n, and with reasoning off it ends in a prefilled empty <think></think>. Reasoning-off is therefore not "omit the block" — a runtime that simply drops <think> will produce a model that opens one and never closes it.

The template also defaults truncate_history_thinking=True, so reasoning traces in earlier assistant turns are dropped when a new user turn arrives.

Tool calling

Tools are rendered into the system prompt and calls come back as XML, not JSON:

<tool_call>
<function=get_weather>
<parameter=city>
Santa Clara
</parameter>
</function>
</tool_call>

Tool results are passed back with role: "tool"; consecutive results are coalesced into a single user turn wrapped in <tool_response>.

Modality

Text only. Verified against the weights, not the config: the checkpoint contains no vision, audio, or video tensors. This is not the Omni or Audex line.

Usage

from mlx_lm import load, generate
from mlx_lm.sample_utils import make_sampler

model, tokenizer = load("OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M")
prompt = tokenizer.apply_chat_template(
    [{"role": "user", "content": "Explain speculative decoding in two sentences."}],
    add_generation_prompt=True, tokenize=False, enable_thinking=True,
)
print(generate(model, tokenizer, prompt=prompt, max_tokens=512,
               sampler=make_sampler(temp=1.0, top_p=0.95), verbose=True))

Multi-Token Prediction

The upstream checkpoint ships a native MTP head and it is retained in this bundle (mtp.*, DeepSeek-V3 shape, sharing the embeddings and output head). Current MLX runtimes do not decode with it, so it is inert — it costs disk only, not RAM. The bundle metadata reflects this honestly: mtp.artifact_available = true, mtp.runtime_available = false. Presence of these weights is not a claim of active speculative-decoding acceleration.

License

Released under the OpenMDW-1.1 license of the base model. Please review the base model's terms before use.

Converted by Jinho Jang — eric@osaurus.ai — with JANG.

Downloads last month: 20

Safetensors

Model size

5B params

Tensor type

U32

F16

F32

MLX

Hardware compatibility

Quantized

Model tree for OsaurusAI/Nemotron-3.5-Lightning-30B-A3B-JANG_4M

Base model

nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16

Quantized

(57)

this model