Apple Mac Studio (M5 Ultra, 80-core GPU) vs Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell
The Apple Mac Studio (M5 Ultra, 80-core GPU) gives a model 192 GB to work in (an assumed figure), against 96 GB on the Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell. The Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell has 1.5× the memory bandwidth (1,792 against 1,200 GB/s), which sets the ceiling on generation speed. The largest 4-bit model each holds at 8,192 tokens of context: GLM-5.3 Flash on the Apple Mac Studio (M5 Ultra, 80-core GPU), Nemotron 3 Super 120B-A12B on the Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell. The cheapest Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell lists for about $7,110 more than the cheapest Apple Mac Studio (M5 Ultra, 80-core GPU). Rows where the two differ are tinted.
Apple unified memory
Apple Mac Studio (M5 Ultra, 80-core GPU)
The full M5 Ultra Mac Studio. Apple lists a 512GB option for this chip as coming in late October 2026; until it ships, 256GB is the largest orderable memory.
gpt-oss-20b, 4-bit: ~200–357 tok/sestimate
NVIDIA workstation
Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell
A single-socket Dell workstation tower configured with NVIDIA's 96GB RTX PRO 6000 Blackwell card. It is the simplest way to buy 96GB of fast VRAM with a manufacturer warranty, at a large premium over a built PC.
gpt-oss-20b, 4-bit: 287 tok/s
| Specification | Apple Mac Studio (M5 Ultra, 80-core GPU) | Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell |
|---|---|---|
| Kind | Apple unified memory · Apple | NVIDIA workstation · Dell |
| Chip or GPUs | M5 Ultra (36-core CPU, 80-core GPU) | RTX PRO 6000 Blackwell Workstation Edition |
| Memory | 96 GB, 256 GB | 96 GB |
| Memory for models | 192 GB (Default GPU limit)assumption | 96 GB (GPU-usable memory) |
| Other memory figure | 248 GB (Limit raised with sysctl)assumption | none |
| Memory bandwidth | 1,200 GB/s | 1,792 GB/s |
| Price | ~$6,799–$10,799 | ~$13,909 (with 8GB of system RAM; 128GB adds $3,940) |
| Largest at 4-bit, 8k context | GLM-5.3 Flash | Nemotron 3 Super 120B-A12B |
| Software | Metal, MLX | CUDA, Vulkan |
| Power | 480 W | 1500 W |
What each one runs
Every model in the catalog that at least one of the two holds at 4-bit with 8,192 tokens of context. Copper speeds are measured medians; outlined ones are estimates. Rows where only one fits, or where the speed differs by more than a quarter, are tinted.
| Model | Apple Mac Studio (M5 Ultra, 80-core GPU) | Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell |
|---|---|---|
| MiniMax M3428B, 23B active · 243.3 GB file | Fits with raised limitassumed limit~28–51 tok/sestimate | does not fit |
| Qwen3.5 397B-A17B397B, 17B active · 227.3 GB file | Fits with raised limitassumed limit~38–68 tok/sestimate | does not fit |
| GLM-5.3 Flash320B, 18B active · 186.0 GB file | ~36–63 tok/sestimate | does not fit |
| Hy3295B, 21B active · 169.6 GB file | ~31–55 tok/sestimate | does not fit |
| MiniMax M2.7228.7B · 129.3 GB file | – | does not fit |
| Qwen3.8 Flash-Next125B, 6B active · 111.4 GB file | ~70–124 tok/sestimate | does not fit |
| Nemotron 3 Super 120B-A12B120B, 12B active · 81.0 GB file | ~46–82 tok/sestimate | ~51–75 tok/sestimate |
| Qwen3.5 122B-A10B122B, 10B active · 71.3 GB file | ~64–113 tok/sestimate | 98 tok/smedian of 1 run |
| Mistral Medium 3.5 128B128B · 69.8 GB file | ~12–15 tok/sestimate | ~12–18 tok/sestimate |
| GLM-4.5-Air106B, 12B active · 68.0 GB file | ~48–86 tok/sestimate | 101 tok/smedian of 1 run |
| Mistral Small 4 119B (2603)119B, 6.5B active · 67.6 GB file | ~101–179 tok/sestimate | ~112–164 tok/sestimate |
| Llama 4 Scout 17B-16E Instruct109B, 17B active · 60.9 GB file | ~39–70 tok/sestimate | ~44–64 tok/sestimate |
| gpt-oss-120b117B, 5.1B active · 58.5 GB file | 111 tok/smedian of 1 run | 202 tok/smedian of 4 runs |
| Qwen3 Coder Next80B, 3B active · 45.2 GB file | 102 tok/smedian of 2 runs | 91 tok/smedian of 1 run |
| Llama 3.3 70B Instruct70B · 39.6 GB file | ~21–26 tok/sestimate | 32 tok/smedian of 1 run |
| Nemotron 3.5 Lightning 30B-A3B30B, 3B active · 23.7 GB file | ~157–279 tok/sestimate | ~175–256 tok/sestimate |
| Qwen3.6 35B-A3B35B, 3B active · 20.8 GB file | ~209–373 tok/sestimate | ~233–341 tok/sestimate |
| Ornith 1.5 35B-A3B35B, 3B active · 20.2 GB file | ~214–382 tok/sestimate | ~240–350 tok/sestimate |
| Qwen3 Coder 30B-A3B Instruct30.5B, 3.3B active · 17.3 GB file | ~199–354 tok/sestimate | ~222–324 tok/sestimate |
| Gemma 4 31B IT30.7B · 17.1 GB file | ~48–60 tok/sestimate | 62 tok/smedian of 1 run |
| Granite 4.2 30B29.3B · 16.8 GB file | ~49–61 tok/sestimate | ~49–77 tok/sestimate |
| Qwen3.8 27B27B · 16.2 GB file | 50 tok/smedian of 4 runs | ~50–79 tok/sestimate |
| Gemma 4 26B-A4B IT25.2B, 3.8B active · 15.9 GB file | ~155–277 tok/sestimate | 197 tok/smedian of 1 run |
| Qwen3.6 27B27B · 15.7 GB file | ~52–65 tok/sestimate | ~52–82 tok/sestimate |
| Devstral Small 2 24B (2512)24B · 13.3 GB file | ~61–77 tok/sestimate | ~61–96 tok/sestimate |
| gpt-oss-20b21B, 3.6B active · 10.8 GB file | ~200–357 tok/sestimate | 287 tok/smedian of 3 runs |
| Phi-414.7B · 8.3 GB file | ~99–123 tok/sestimate | ~99–155 tok/sestimate |
| Ministral 3 14B Reasoning (2512)13.9B · 7.7 GB file | ~107–133 tok/sestimate | ~107–167 tok/sestimate |
| Gemma 4 12B IT11.95B · 6.6 GB file | ~124–154 tok/sestimate | ~123–194 tok/sestimate |
| Qwen3.5 9B9B · 5.3 GB file | ~155–193 tok/sestimate | ~155–243 tok/sestimate |
| Granite 4.2 8B8.8B · 5.2 GB file | ~159–198 tok/sestimate | ~159–249 tok/sestimate |
| Gemma 4 E4B IT8B · 5.0 GB file | ~290–361 tok/sestimate | ~289–454 tok/sestimate |
| Qwen3 8B8.2B · 4.7 GB file | ~175–218 tok/sestimate | 174 tok/smedian of 1 run |
| Llama 3.1 8B Instruct8B · 4.6 GB file | ~179–223 tok/sestimate | 144 tok/smedian of 2 runs |
| Qwen3.5 4B4B · 2.6 GB file | ~321–400 tok/sestimate | ~321–503 tok/sestimate |
Full pages: Apple Mac Studio (M5 Ultra, 80-core GPU) · Dell Pro Max Tower T2 with NVIDIA RTX PRO 6000 Blackwell · check a different quant or context length