Check my GPU

on

Can the Apple M3 Ultra (512 GB) run InternVL3.5 241B-A28B?

81 ptsGreat (A) · 81 of 100 points: Model quality 36.8/45, Memory fit 25.0/25, Speed 9.3/15, Memory use 10.8/15

Yes, it runs well

Runs well. At Q4_K_M it needs about 180 GB of 384 GB usable, with an estimated speed of 15.5 tok/s.

Best quant
Q4_K_M
Generation speed
15.5 tok/s
Prompt processing
629 tok/s
First token (1K prompt)
1.6 s
Largest safe context
40K

Download InternVL3.5 241B-A28B

Quantized files · GGUF

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

huggingface.co/mradermacher/InternVL3_5-241B-A28B-GGUF (opens in a new tab)

ollama run hf.co/mradermacher/InternVL3_5-241B-A28B-GGUF

Original weights · Hugging Face

Full-precision files from Shanghai AI Lab, for fine-tuning, vLLM, Transformers.

huggingface.co/OpenGVLab/InternVL3_5-241B-A28B (opens in a new tab)

No sign-in needed.

Links checked 4 hours ago against Hugging Face, Ollama and GitHub.

How the A 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.36.8 / 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.9.3 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.10.8 / 15
Total81.9 / 10081 ptsGreat (A) · 81 of 100 points: Model quality 36.8/45, Memory fit 25.0/25, Speed 9.3/15, Memory use 10.8/15

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

Every quantization on the Apple M3 Ultra (512 GB)

Quant Memory needed Available Speed Fit Grade · points Download
Q2_K 129 GB 384 GB 25.4 tok/s Runs well 71 ptsGood (B) · 71 of 100 points: Model quality 27.1/45, Memory fit 25.0/25, Speed 12.0/15, Memory use 7.8/15 GGUF files (opens in a new tab)
Q3_K_M 151 GB 384 GB 20.0 tok/s Runs well 77 ptsGreat (A) · 77 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 9.1/15 GGUF files (opens in a new tab)
Q4_K_M 180 GB 384 GB 15.5 tok/s Runs well 81 ptsGreat (A) · 81 of 100 points: Model quality 36.8/45, Memory fit 25.0/25, Speed 9.3/15, Memory use 10.8/15 GGUF files (opens in a new tab)
Q5_K_M 204 GB 384 GB 12.8 tok/s Runs well 83 ptsGreat (A) · 83 of 100 points: Model quality 37.5/45, Memory fit 25.0/25, Speed 8.5/15, Memory use 12.2/15 GGUF files (opens in a new tab)
Q6_K 230 GB 384 GB 10.7 tok/s Runs well 84 ptsGreat (A) · 84 of 100 points: Model quality 38.3/45, Memory fit 25.0/25, Speed 7.8/15, Memory use 13.8/15 GGUF files (opens in a new tab)
Q8_0 288 GB 384 GB 7.9 tok/s Runs well 85 ptsExcellent (S) · 85 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 6.7/15, Memory use 15.0/15 GGUF files (opens in a new tab)
F16 514 GB 384 GB — Won't run 11 ptsWon't run (F) · 11 of 100 points: Model quality 11.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 Original weights (opens in a new tab)

Memory figures assume 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.