Skip to content
Local AI Wizard
Your machine
Theme

Nemotron 3 Nano 30B A3B

NVIDIA · nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16

Q4 file 24.6 GB (Q4_K_M) · runs well from 32 GB · Weights license: Publisher terms: NVIDIA Nemotron Open Model License

LMArena · text1,314 ± 5Rank 273 · release 2026-10-22 · its own scale
Memory needed~30 GBQ4 weights + working room, before the machine's reserve
Parameters30 Bmanufacturer specification · 3.5B active
Context262,144 tokensPublished for this model

Quality

LMArena · text · rating
1,314 ± 5Confidence interval 1,308–1,319 · 15,765 votes · rank 273 of the release · Arena name carries (BF16), the same BF16 weightsLMArena 2026-10-22 · CC BY 4.0

No result in LiveBench 2026-06-25 Not in LiveBench 2026-06-25

Reported by the publisher

29 results · NVIDIA's model card · read 2026-10-25
MMLU-Pro
78.3General Knowledge
AIME25 (no tools)
89.1Reasoning
AIME25 (with tools)
99.2Reasoning
GPQA (no tools)
73.0Reasoning
GPQA (with tools)
75.0Reasoning
LiveCodeBench (v6 2025-08–2025-05)
68.3Reasoning
SciCode (subtask)
33.3Reasoning
HLE (no tools)
10.6Reasoning
Every reported result29
HLE (with tools)
15.5Reasoning
MiniF2F pass@1
50.0Reasoning
MiniF2F pass@32
79.9Reasoning
Terminal Bench (hard subset)
8.5Agentic
SWE-Bench (OpenHands)
38.8Agentic
TauBench V2 (Airline)
48.0Agentic
TauBench V2 (Retail)
56.9Agentic
TauBench V2 (Telecom)
42.2Agentic
TauBench V2 (Average)
49.0Agentic
BFCL v4
53.8Agentic
IFBench (prompt)
71.5Chat & Instruction Following
Scale AI Multi Challenge
38.5Chat & Instruction Following
Arena-Hard-V2 (Hard Prompt)
72.1Chat & Instruction Following
Arena-Hard-V2 (Creative Writing)
63.2Chat & Instruction Following
Arena-Hard-V2 (Average)
67.7Chat & Instruction Following
AA-LCR
35.9Long Context
RULER-100@256k
92.9Long Context
RULER-100@512k
91.3Long Context
RULER-100@1M
86.3Long Context
MMLU-ProX (avg over langs)
59.5Multilingual
WMT24++ (en-\>xx)
86.2Multilingual

Values as printed on the card, with the publisher's own settings (card section: License/Terms of Use › Reasoning Benchmark Evaluations); column “NVIDIA-Nemotron-3-Nano-30B-A3B-BF16”. Not comparable across publishers and not used in any ranking. Model card

Run it

Run it

ollama run hf.co/unsloth/Nemotron-3-Nano-30B-A3B-GGUF:Q4_K_M
  1. Install Ollama
  2. Run the command above
  3. Ask: Explain what a context window is, in two sentences.

Runs on

Fast memory holds it 30

The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.

Show all 30

Within memory, outside the fast part 46

Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.

Show all 46

One memory tier short 3

The same machines with the next memory size up would hold it in fast memory.

Nemotron 3 Nano 30B A3B is a text model from NVIDIA with 30B parameters, 3.5B of them active per token (mixture of experts). It was released on December 15, 2025 and reads up to 262,144 tokens of context. Its Q4 weights plus working room come to about 30 GB before the machine's reserve; the smallest reference machines whose fast memory holds that are Mac mini M5 Pro · 64 GB unified (~38 GB with its reserve) and RTX 5090 · 32 GB VRAM (~32 GB with its reserve). We collected 26 downloadable GGUF and MLX files from 2 publisher sources; the most downloaded GGUF source is unsloth's. It has no result in LiveBench 2026-06-25. Its LMArena text rating is 1,314 ± 5 (release 2026-10-22), a separate human-preference scale.

Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Each card opens its hardware profile. Not a measured fit or speed.

On the reference machines

Nemotron 3 Nano 30B A3B on the reference machines
Where Q4 sits
Mac mini M2 · 8 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~32 GBLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
Mac mini M4 · 16 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~34 GBLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
Mac mini M4 · 24 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~36 GBLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
Mac mini M4 · 32 GB unifiedLarger than its memoryIQ4_XSGGUF · 18.2 GB · needs ~30 GB262,144 tokens~38 GBLarger than its memoryIQ4_XSGGUF · 18.2 GB · needs ~30 GB262,144 tokens
Mac mini M5 Pro · 64 GB unifiedFast memory holds itQ8_0GGUF · 33.6 GB · needs ~49 GB262,144 tokens~38 GBFast memory holds itQ8_0GGUF · 33.6 GB · needs ~49 GB262,144 tokens
Mac Studio M5 Max · 128 GB unifiedFast memory holds itQ8_0GGUF · 33.6 GB · needs ~49 GB262,144 tokens~38 GBFast memory holds itQ8_0GGUF · 33.6 GB · needs ~49 GB262,144 tokens
Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unifiedFast memory holds itQ8_0GGUF · 33.6 GB · needs ~49 GB262,144 tokens~38 GBFast memory holds itQ8_0GGUF · 33.6 GB · needs ~49 GB262,144 tokens
RTX 4060 · 8 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~32 GBIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
RTX 3060 · 12 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~32 GBIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
RTX 5080 · 16 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~32 GBIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
RTX 4090 · 24 GB VRAMIn memory, outside the fast partIQ4_XSGGUF · 18.2 GB · needs ~24 GB16,384 tokens~32 GBIn memory, outside the fast partIQ4_XSGGUF · 18.2 GB · needs ~24 GB16,384 tokens
RTX 5090 · 32 GB VRAMFast memory holds itQ4_K_MGGUF · 24.6 GB · needs ~32 GB65,536 tokens~32 GBFast memory holds itQ4_K_MGGUF · 24.6 GB · needs ~32 GB65,536 tokens
DGX H100 · 8 × 80 GBFast memory holds itQ8_0GGUF · 33.6 GB · needs ~57 GB262,144 tokens~46 GBFast memory holds itQ8_0GGUF · 33.6 GB · needs ~57 GB262,144 tokens

The file column uses the site's quant ladder (Q8_0 to Q3_K_M, MLX 8 to 4-bit on a Mac): the largest collected file whose size × 1.2 plus the context cache and the machine's reserve fits the fast memory. Planning estimates, not measured runs.

Plan another quantization or context

Consider it for

Compare Nemotron 3 Nano 30B A3B using its publisher card. No overall score appears in the LiveBench 2026-06-25 snapshot.

Check first

The Q4 weights plus working room come to about 30 GB from 30B stated parameters, before the machine's reserve (Runs on adds it per machine). It is not a measured minimum; context, batch size, runtime and quantization change actual memory use.

Versions to download

Files you can download for this checkpoint. The section opens on the most downloaded collected GGUF source, unsloth, at Q4_K_M. On a Mac it opens on the most downloaded MLX source. Sizes are the Hub's published bytes. What is Q4?

unsloth

unsloth/Nemotron-3-Nano-30B-A3B-GGUF · Nemotron-3-Nano-30B-A3B-Q4_K_M.gguf · 24.6 GB

Run this file

ollama run hf.co/unsloth/Nemotron-3-Nano-30B-A3B-GGUF:Q4_K_M
llama-server -hf unsloth/Nemotron-3-Nano-30B-A3B-GGUF:Q4_K_M
lms get https://huggingface.co/unsloth/Nemotron-3-Nano-30B-A3B-GGUF
  • mlx-lm: mlx-lm runs MLX repositories; pick the MLX version on a Mac

Quality scores on this page belong to the original checkpoint; these files are not scored or ranked. File sizes read 2026-10-23 · download counts read 2026-10-17 · commands: Hugging Face Hub docs: Use Ollama with any GGUF model · llama.cpp server README (-hf <user>/<model>[:quant]) · LM Studio docs: lms get with a Hugging Face URL · mlx-lm README (mlx_lm.generate --model)

Fine-tunes, adapters and other versions73 fine-tunes
Quantized versions
72
Fine-tunes
73
Adapters · merges
285 · 0

Browse them on the Hugging Face model tree · read 2026-10-23

Uncensored and abliterated versions1

Some community uploads remove the model's safety refusals. Quality may drop, the original publisher does not support them, claims about them are the uploader's, and the original license still applies. This site never recommends or ranks them.

1 derivative repositories have “uncensored” or “abliterated” in their name (read 2026-10-23).

Measured speed

No published measurement for this checkpoint

Tokens per second are shown only when a source publishes them with hardware, build and setup. They are never estimated.

How to run

The memory needed above is the Q4 load before a machine's reserve. Its Q4_K_M file (24.6 GB) is heavier than the parameter formula gives, so the memory needed is that file × 1.2. Formula: parameters × 0.625 bytes, plus 20% working allowance; each machine then adds its system reserve: 25% of unified or system memory, 2 to 8 GB, or 2 GB per GPU on dedicated cards. This is neither measured VRAM nor checkpoint download size.

Q4 planning estimate only. Check checkpoint format, runtime support, KV cache, context, batch size and offload before choosing hardware.

Evidence level: Generic estimate only; no execution reproduced by this site.

Publisher sampling defaults
temperature 1 · top_p 1generation_config.json · read 2026-10-23source
Chat template markers
tool calls · thinkingmarkers found in the template, not a test of the feature · read 2026-10-23source
Before downloading
  1. Check the exact checkpoint and license linked here.
  2. Confirm a compatible runtime and precision or quantization for your platform.
  3. Allow for context, batch size and any extra media components beyond the weights.

Technical details

What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.

Task and modality
language and reasoning · text
Count and basis
30B · manufacturer specification · 3.5B activenvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 model card (Hugging Face) · read 2026-10-17source
Context
262,144 tokens published for this modelnvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 config.json (Hugging Face) · read 2026-10-17source
Architecture
MoEnvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 model card (Hugging Face) · read 2026-10-17source
Released
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 model card (Hugging Face) · read 2026-10-17source
Runtime and precision
No local execution verified.
Languages
en, es, fr, de, ja, itModel card metadata · read 2026-10-25source
Commercial use
License text not read: nvidia.com was unreachable on 2026-10-18 and the repositories ship no copy
Every stored field12 fields
Revision checked
bf77c3174f68ad409e1c2aa60daeb46e32d1c606
Architecture
NemotronHForCausalLM
Library
transformers
Hub task tag
text-generation
Formats
safetensors
Access
Open download
Publisher kind
Official
Repository created
2025-12-04creation is not the model's release
Repository updated
2026-08-24a commit is not a release date
Hub downloads (30 days) · likes
1,438,724 · 835this repository only · read 2026-10-25
Read from the Hub page
transformers class, task tags, license nameHub overview and file listing · read 2026-10-18
Catalog state
Hub config or weights found; no inventory benchmark link

Hybrid Mamba-2 and attention mixture-of-experts: only 6 of 52 layers use attention. The linked repository is the BF16 reference checkpoint; the generic Q4 estimate is not a measured requirement.

Sources

Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.

Catalog reviewed: 2026-10-17. Benchmark release: 2026-06-25; snapshot collected: 2026-09-26.