Check my GPU

on

Can the Apple M3 Ultra (256 GB) run Qwen3 Next 80B-A3B Instruct?

90 ptsExcellent (S) · 90 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.3/15

Yes, it runs well

Runs well. At Q8_0 it needs about 102 GB of 192 GB usable, with an estimated speed of 73.5 tok/s.

Best quant
Q8_0
Generation speed
73.5 tok/s
Prompt processing
5,869 tok/s
First token (1K prompt)
0.3 s
Largest safe context
256K

Download Qwen3 Next 80B-A3B Instruct

Recommended for the Apple M3 Ultra (256 GB): Q8_0. Every other quantization has its own download link in the table below.

Quickest · Ollama

One command downloads a ready-to-run version and starts a chat.

ollama run qwen3-next:80b-a3b-instruct-q8_0

Get Ollama at ollama.com/download (opens in a new tab) · model page (opens in a new tab)

Quantized files · GGUF

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

Download the Q8_0 file (opens in a new tab)

huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct-GGUF (opens in a new tab)

ollama run hf.co/Qwen/Qwen3-Next-80B-A3B-Instruct-GGUF:Q8_0

Original weights · Hugging Face

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

huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct (opens in a new tab)

No sign-in needed.

Official repo · GitHub

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

github.com/QwenLM/Qwen3 (opens in a new tab)

git clone https://github.com/QwenLM/Qwen3

Qwen’s main repo, renamed from Qwen2.5 to Qwen3; it still documents Qwen2.5.

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

How the S grade is scored

Every result gets up to 100 points from four parts. The points decide the letter grade.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.37.8 / 45
Memory fitFull marks when everything sits in GPU or unified memory; fewer when part spills to system RAM.25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.12.3 / 15
Total90.1 / 10090 ptsExcellent (S) · 90 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.3/15

Grade scale: S 85+ · A 72+ · B 58+ · C 42+ · D 25+ · F under 25 points.

Every quantization on the Apple M3 Ultra (256 GB)

Quant Memory needed Available Speed Fit Grade · points Download
Q2_K 48.7 GB 192 GB 237 tok/s Runs well 72 ptsGreat (A) · 72 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15 GGUF files (opens in a new tab)
Q3_K_M 56.0 GB 192 GB 187 tok/s Runs well 78 ptsGreat (A) · 78 of 100 points: Model quality 31.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15 GGUF files (opens in a new tab)
Q4_K_M 65.8 GB 192 GB 144 tok/s Runs well 83 ptsGreat (A) · 83 of 100 points: Model quality 35.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.9/15 GGUF (opens in a new tab)
Q5_K_M 73.9 GB 192 GB 119 tok/s Runs well 85 ptsExcellent (S) · 85 of 100 points: Model quality 36.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.9/15 GGUF (opens in a new tab)
Q6_K 82.8 GB 192 GB 100 tok/s Runs well 87 ptsExcellent (S) · 87 of 100 points: Model quality 37.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.0/15 GGUF (opens in a new tab)
Q8_0 102 GB 192 GB 73.5 tok/s Runs well 90 ptsExcellent (S) · 90 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.3/15 GGUF (opens in a new tab)
F16 179 GB 192 GB 33.9 tok/s Tight fit 90 ptsExcellent (S) · 90 of 100 points: Model quality 37.8/45, Memory fit 24.0/25, Speed 13.8/15, Memory use 15.0/15 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

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.