on

Can the Apple M5 Ultra (96 GB) run Qwen1.5 MoE A2.7B Chat?

24Basic quality (D) · 24 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking.

Yes, it runs great

Runs great. At F16 it needs about 36.6 GB of 72.0 GB usable, with an estimated speed of 74.4 tok/s.

Best quant
F16
Generation speed
74.4 tok/s
Prompt processing
8,657 tok/s
First token (1K prompt)
0.3 s
Largest safe context
8K

Best app to run it on the Apple M5 Ultra (96 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX · use the MLX build

  • Runs the MLX build, typically 35–90% faster than GGUF for a model with 2.7B active parameters on Apple silicon.
  • Point-and-click: search, download and chat in one window.

Open Qwen1.5 MoE A2.7B Chat in LM Studio

  • MLX LM Runs the MLX build, typically 35–90% faster than GGUF for a model with 2.7B active parameters on Apple silicon.
  • Unsloth Desktop Runs the MLX build, typically 35–90% faster than GGUF for a model with 2.7B active parameters on Apple silicon.

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 Qwen1.5 MoE A2.7B Chat

Original weights · Hugging Face

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

huggingface.co/Qwen/Qwen1.5-MoE-A2.7B-Chat (opens in a new tab)

No sign-in needed.

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 24Basic quality (D) · 24 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking.
How capable the model itself is at F16, 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 Runs great
How well it runs on the Apple M5 Ultra (96 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.10.8 / 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.15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.11.7 / 15
Match score62.5 / 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 Apple M5 Ultra (96 GB)

Quant Memory needed Available Speed Runs Quality Download
Q2_K 13.7 GB 72.0 GB 226 tok/s Runs great 16Basic quality (D) · 16 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q3_K_M 15.0 GB 72.0 GB 205 tok/s Runs great 20Basic quality (D) · 20 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q4_K_M 16.7 GB 72.0 GB 182 tok/s Runs great 22Basic quality (D) · 22 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q5_K_M 18.1 GB 72.0 GB 165 tok/s Runs great 23Basic quality (D) · 23 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q6_K 19.7 GB 72.0 GB 150 tok/s Runs great 23Basic quality (D) · 23 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q8_0 23.1 GB 72.0 GB 125 tok/s Runs great 24Basic quality (D) · 24 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
F16 36.6 GB 72.0 GB 74.4 tok/s Runs great 24Basic quality (D) · 24 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. 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

Quick answers

How much memory does Qwen1.5 MoE A2.7B Chat need on the M5 Ultra 96GB?

At F16 it needs about 36.6 GB. The M5 Ultra 96GB has 72.0 GB usable.

How fast is Qwen1.5 MoE A2.7B Chat on the M5 Ultra 96GB?

About 74.4 tok/s at F16, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Qwen1.5 MoE A2.7B Chat should I use on the M5 Ultra 96GB?

F16. 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.

Was this page helpful?