Llama 3.3 Nemotron Super 49B v1.5 on AMD Ryzen AI Max+ 395 (128 GB)

Can the AMD Ryzen AI Max+ 395 (128 GB) run Llama 3.3 Nemotron Super 49B v1.5?

Yes, but slowly

Slow. At Q6_K it needs about 51.9 GB of 96.0 GB usable, with an estimated speed of 3.8 tok/s.

Best quant: Q6_K
Generation speed: 3.8 tok/s
Prompt processing: 738 tok/s
First token (1K prompt): 1.4 s
Largest safe context: 128K

Best app to run it on the AMD Ryzen AI Max+ 395 (128 GB)

On Windows

Lemonade

Desktop app · runs on llama.cpp, ONNX Runtime GenAI (Ryzen AI NPU)

Built for Ryzen AI and Strix Halo, including NPU + iGPU modes on Windows.

lemonade pull unsloth/Llama-3_3-Nemotron-Super-49B-v1_5-GGUF:Q6_K

LM Studio Point-and-click: search, download and chat in one window.
llama.cpp Usually the first to support new model architectures.

On Linux

Lemonade

Desktop app · runs on llama.cpp, ONNX Runtime GenAI (Ryzen AI NPU)

Built for Ryzen AI and Strix Halo, including NPU + iGPU modes on Windows.

lemonade pull unsloth/Llama-3_3-Nemotron-Super-49B-v1_5-GGUF:Q6_K

LM Studio Point-and-click: search, download and chat in one window.
llama.cpp Usually the first to support new model architectures.

Want an agent? Llama 3.3 Nemotron Super 49B v1.5 supports tool use, so you can pair a llama.cpp or MLX LM server with one of these. Apps that run on this computer come first.

Coding agents

Personal assistants

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Llama 3.3 Nemotron Super 49B v1.5

Recommended for the AMD Ryzen AI Max+ 395 (128 GB): Q6_K. Every other quantization has its own download link in the table below.

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

Download the Q6_K file (opens in a new tab)

huggingface.co/unsloth/Llama-3_3-Nemotron-Super-49B-v1_5-GGUF (opens in a new tab)

ollama run hf.co/unsloth/Llama-3_3-Nemotron-Super-49B-v1_5-GGUF:Q6_K

Original weights · Hugging Face

Full-precision files from NVIDIA, for fine-tuning, vLLM, Transformers.

huggingface.co/nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from NVIDIA. The weights themselves download from Hugging Face.

github.com/NVIDIA-NeMo/Nemotron (opens in a new tab)

NVIDIA’s Nemotron repo with recipes for the whole family.

git clone https://github.com/NVIDIA-NeMo/Nemotron

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 79Great quality (A) · 79 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q6_K, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Slow: How well it runs on the AMD Ryzen AI Max+ 395 (128 GB): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	35.6 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.	25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	4.6 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	12.5 / 15
Match score	77.7 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the AMD Ryzen AI Max+ 395 (128 GB)

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	30.9 GB	96.0 GB	8.9 tok/s	Slow	56Good quality (B) · 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q3_K_M`	35.4 GB	96.0 GB	7.0 tok/s	Slow	67Good quality (B) · 67 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q4_K_M`	41.4 GB	96.0 GB	5.4 tok/s	Slow	76Great quality (A) · 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	46.4 GB	96.0 GB	4.5 tok/s	Slow	77Great quality (A) · 77 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	51.9 GB	96.0 GB	3.8 tok/s	Slow	79Great quality (A) · 79 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	63.8 GB	96.0 GB	2.8 tok/s	Very slow	80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`F16`	111 GB	96.0 GB	1.1 tok/s	Very slowfar over GPU memory limit	80Great quality (A) · 80 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab) Original weights (opens in a new tab)

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

77Great quality (A) · 77 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q5_K_M · 36.4 tok/s · Runs great
67Good quality (B) · 67 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Q3_K_M · 57.0 tok/s · Runs well (tight fit)
76Great quality (A) · 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Max (64 GB) Q4_K_M · 11.6 tok/s · Runs well
67Good quality (B) · 67 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. Apple M1 Max (64 GB) Q3_K_M · 11.0 tok/s · Runs well
76Great quality (A) · 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Max (128 GB) Q4_K_M · 11.6 tok/s · Runs well
76Great quality (A) · 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. Apple M2 Ultra (192 GB) Q4_K_M · 17.0 tok/s · Runs well

Quick answers

How much memory does Llama 3.3 Nemotron Super 49B v1.5 need on the AMD Ryzen AI Max+ 395 128GB?

At Q6_K it needs about 51.9 GB. The AMD Ryzen AI Max+ 395 128GB has 96.0 GB usable.

How fast is Llama 3.3 Nemotron Super 49B v1.5 on the AMD Ryzen AI Max+ 395 128GB?

About 3.8 tok/s at Q6_K, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Llama 3.3 Nemotron Super 49B v1.5 should I use on the AMD Ryzen AI Max+ 395 128GB?

Q6_K. Of the versions that fit, it is the one that runs best; when two run equally well, the check prefers the higher-rated one.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.