Check my GPU

on

Can the Apple M2 Pro (32 GB) run DeepSeek MoE 16B Chat?

64 ptsGood (B) · 64 of 100 points: Model quality 11.1/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 14.3/15

Yes, it runs well

Runs well. At Q4_K_M it needs about 13.2 GB of 21.4 GB usable, with an estimated speed of 37.8 tok/s.

Best quant
Q4_K_M
Generation speed
37.8 tok/s
Prompt processing
1,506 tok/s
First token (1K prompt)
0.7 s
Largest safe context
4K

Download DeepSeek MoE 16B Chat

Recommended for the Apple M2 Pro (32 GB): Q4_K_M. 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 Q4_K_M file (opens in a new tab)

huggingface.co/mradermacher/deepseek-moe-16b-chat-i1-GGUF (opens in a new tab)

ollama run hf.co/mradermacher/deepseek-moe-16b-chat-i1-GGUF:Q4_K_M

Original weights · Hugging Face

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

huggingface.co/deepseek-ai/deepseek-moe-16b-chat (opens in a new tab)

No sign-in needed.

Links checked 6 days ago against Hugging Face, Ollama and GitHub.

How the B 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.11.1 / 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.14.6 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.14.3 / 15
Total64.9 / 10064 ptsGood (B) · 64 of 100 points: Model quality 11.1/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 14.3/15

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

Every quantization on the Apple M2 Pro (32 GB)

Quant Memory needed Available Speed Fit Grade · points Download
Q2_K 9.8 GB 21.4 GB 62.0 tok/s Runs well 58 ptsGood (B) · 58 of 100 points: Model quality 8.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.6/15 GGUF (opens in a new tab)
Q3_K_M 11.3 GB 21.4 GB 48.8 tok/s Runs well 61 ptsGood (B) · 61 of 100 points: Model quality 9.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.1/15 GGUF (opens in a new tab)
Q4_K_M 13.2 GB 21.4 GB 37.8 tok/s Runs well 64 ptsGood (B) · 64 of 100 points: Model quality 11.1/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 14.3/15 GGUF (opens in a new tab)
Q5_K_M 14.9 GB 21.4 GB 31.2 tok/s Runs well 64 ptsGood (B) · 64 of 100 points: Model quality 11.3/45, Memory fit 25.0/25, Speed 13.2/15, Memory use 15.0/15 GGUF (opens in a new tab)
Q6_K 16.7 GB 21.4 GB 26.2 tok/s Runs well 63 ptsGood (B) · 63 of 100 points: Model quality 11.6/45, Memory fit 25.0/25, Speed 12.1/15, Memory use 15.0/15 GGUF (opens in a new tab)
Q8_0 20.6 GB 21.4 GB 19.2 tok/s Tight fit 61 ptsGood (B) · 61 of 100 points: Model quality 11.7/45, Memory fit 24.0/25, Speed 10.4/15, Memory use 15.0/15 GGUF files (opens in a new tab)
F16 36.0 GB 21.4 GB — Won't run 3 ptsWon't run (F) · 3 of 100 points: Model quality 3.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 Original weights (opens in a new tab)

Memory figures assume 4K 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.