NVIDIA DGX Spark vs Mac Studio M5 Max 128 GB
DGX Spark · 128 GB unified and Mac Studio M5 Max · 128 GB unified side by side for local AI.
Facts from both hardware pages
| Specification | NVIDIA DGX Spark | Mac Studio M5 Max 128 GB |
|---|---|---|
| Memory | 128 GB unified memoryShared by CPU and GPU · source | 128 GB unified memoryShared by CPU and GPU · source |
| Memory bandwidth | 273 GB/sManufacturer spec · not model speed · source | 614 GB/sManufacturer spec · not model speed · source |
| Text · run well | 193+1 tight fit · within 128 GB | 191+2 tight fit · 1 run slowly · within 104 GB |
| Vision · run well | 150within 128 GB | 150within 104 GB |
| Image · within memory | 77within 128 GB · documented paths and minimums | 77within 104 GB · documented paths and minimums |
| Video · within memory | 73within 128 GB · documented paths and minimums | 72within 104 GB · documented paths and minimums |
| Audio · within memory | 134within 128 GB · documented paths and minimums | 132within 104 GB · documented paths and minimums |
| Embeddings · within memory | 91within 128 GB · 8-bit estimate | 91within 104 GB · 8-bit estimate |
| Published speed runs | 1413 on this machine · 1 naming only the chip | 11 on this machine |
| Launch price (MSRP) | USD 3,999US · dated 2025-10-15 · not a tracked offer · source | No launch price on file |
| Where to buy | NVIDIA Marketplacethis configuration · US · recorded 2026-10-18 · store link | Apple Storethis configuration · US · recorded 2026-10-18 · store link |
Counts use the same memory estimate as each hardware page: text and vision run well within the fast memory (tight fits counted apart); image, video and audio use documented paths and publisher minimums. Formulas
Top text models each runs well
The first 5 rows of each hardware page's Text list, in its order: LiveBench score, then reviewed models, then Hub downloads. Memory is the planning estimate on that machine.
| Rank | NVIDIA DGX Spark | Mac Studio M5 Max 128 GB |
|---|---|---|
| #1 | Smaug MiniAbacus AI · LiveBench 76.9 · ~29 GB | Smaug MiniAbacus AI · LiveBench 76.9 · ~29 GB |
| #2 | Qwen3.8 27BAlibaba · LiveBench 75.3 · ~29 GB | Qwen3.8 27BAlibaba · LiveBench 75.3 · ~29 GB |
| #3 | Qwen3.6 27BAlibaba · LiveBench 64 · ~29 GB | Qwen3.6 27BAlibaba · LiveBench 64 · ~29 GB |
| #4 | Qwen3 14BAlibaba · ~20 GB | Qwen3 14BAlibaba · ~20 GB |
| #5 | Gemma 4 31BGoogle · ~32 GB | Gemma 4 31BGoogle · ~32 GB |
Measured speed
Published output tokens per second, as measured. A speed is compared only when the same checkpoint, quantization, runtime and lengths ran on both machines themselves; same-chip references are not compared.
No published test ran the same model and settings on both machines.
All published measurements · NVIDIA DGX Spark
| gpt-oss 20B MXFP4 MoE | 83.43 ± 0.59 | MXFP4 MoE | CUDA | llama.cpp bench |
|---|---|---|---|---|
| gemma-3-4b-it-qat Q4_0 | 81.05 ± 0.20 | Q4_0 | llama.cpp CUDA | llama.cpp benches/dgx-spark · maintainer result |
| Qwen3.5 35B A3B MXFP4 | 63.13 | MXFP4 (MoE; 35B total / 3B active) | llama.cpp CUDA (SM121a, source build) | nabe2030/dense-27b-31b-dgx-spark (GitHub) · community benchmark |
| Qwen3 Coder 30B A3B Q8_0 | 61.06 ± 0.23 | Q8_0 | CUDA | llama.cpp bench |
| Llama 2 7B Q4_0 | 60.39 ± 0.05 | Q4_0 | Vulkan | llama.cpp Vulkan scoreboard · community result |
| gpt-oss-120b MXFP4 MoE | 58.72 ± 0.20 | MXFP4 MoE | llama.cpp CUDA | llama.cpp benches/dgx-spark · maintainer result |
| Nemotron-3-Nano-4B Q8_0 | 52.85 ± 0.12 | Q8_0 | llama.cpp CUDA | llama.cpp benches/nemotron · maintainer result |
| GLM-4.7-Flash Q8_0 | 48.68 ± 0.12 | Q8_0 | llama.cpp CUDA | llama.cpp benches/dgx-spark · maintainer result |
| Qwen3.8 27B NVFP4 + DSpark | 36.6 | NVFP4 + DSpark | SGLang | NVIDIA Developer Forum |
| Qwen2.5-Coder-7B-Q8_0 | 29.43 ± 0.02 | Q8_0 | llama.cpp CUDA | llama.cpp benches/dgx-spark · maintainer result |
| Nemotron 3 Super 120B A12B Q4_K | 19.94 ± 0.16 | Q4_K (reported as 'Q4_K' by llama-bench; GGUF from ggml-org) | llama.cpp CUDA | llama.cpp benches/nemotron · DGX Spark |
| Qwen3.8 27B NVFP4 + MTP | 18.5 | NVFP4 + MTP (3 speculative tokens) | vLLM official | NVIDIA Developer Forum |
| Qwen3.6 27B Q4_K_M | 11.85 ± 0.01 | Q4_K_M (15.65 GiB) | llama.cpp CUDA (SM121a, source build) | nabe2030/dense-27b-31b-dgx-spark (GitHub) · community benchmark |
| Qwen3.6 27B BF16 | 4.54 ± 0.00 | BF16 (50.10 GiB) | llama.cpp CUDA (SM121a, source build) | nabe2030/dense-27b-31b-dgx-spark (GitHub) · community benchmark |
All published measurements · Mac Studio M5 Max 128 GB
| Llama 7B Q4_0 (llama-bench reference model) | 119.92 | Q4_0 | Metal | llama.cpp Apple Silicon summary · community result |
|---|
Verdict
Both have 128 GB unified memory. On Mac Studio M5 Max 128 GB models are planned within 104 GB, the default macOS GPU share. By the memory estimate, 193 text models run well on NVIDIA DGX Spark and 191 on Mac Studio M5 Max 128 GB. Mac Studio M5 Max 128 GB lists 614 GB/s of memory bandwidth, 2.2× the 273 GB/s of NVIDIA DGX Spark (a manufacturer spec, not a measured speed). No published test ran the same model and settings on both, so their speeds are not compared here. Only NVIDIA DGX Spark has a launch price on file: USD 3,999 (2025-10-15). Choose NVIDIA DGX Spark when model size is the limit (more text models run well); no shared measurement compares their speed.
Limits and dates
Model counts are memory estimates, not speed measurements or guaranteed fits; verify runtime and checkpoint support. Bandwidth is the manufacturer's figure. Prices are launch or list prices with their dates, not current offers.
Page updated · Model catalog checked · LiveBench 2026-06-25