Where the numbers come from. Mac figures come from the LLMCheck index, the same per-Mac figures as its Best by Mac pages, each with its provenance mark: a filled square is a published figure, a hollow circle a community run, a diamond an estimate. The DGX Spark and Strix Halo figures are not in the index, which covers Apple Silicon only. They are quoted from the sources linked beside them, with the runtime and date. LLMCheck runs no tests of its own. The machines ran different runtimes and builds, so treat gaps of 10% or less as ties.

The short answer

Memory, bandwidth and price

Mac Studio M5 MaxNVIDIA DGX SparkAMD Strix Halo (Ryzen AI Max+ 395)
Memory128 GB unified128 GB LPDDR5x, unified (a 64 GB version is sold through OEMs)Up to 128 GB LPDDR5X-8000. The GPU gets 96 GB on Windows (AMD) and more on Linux
Memory bandwidth614 GB/s (Apple)273 GB/s (NVIDIA)About 256 GB/s: 256-bit at 8,000 MT/s. AMD publishes no figure
GPU40-core Apple GPUBlackwell with 5th-gen Tensor Cores; NVIDIA claims up to 1 PFLOP of FP4, with sparsityRadeon 8060S, 40 compute units
Operating systemmacOSDGX OS (based on Ubuntu)Windows or Linux
Price$5,099 (Apple list price, checked 14 Nov 2026). 128 GB MacBook Pro: $6,699$3,999 at launch (Oct 2025), then $4,699 from Feb 2026 (NVIDIA). The Register reported $6,950 on 2 Oct 2026; NVIDIA’s own pages showed no price on 18 NovFramework Desktop: $2,459 (Jan 2026), $3,449 by Sep 2026, out of stock on 18 Nov. GMKtec EVO-X2: $3,649 with 2 TB, about $3,499 with 1 TB (Jul 2026). AMD Ryzen AI Halo: $3,999.99 (Jun 2026)

Sources:

Dell’s GB10 workstation (128 GB, 2 TB) was $7,901 on Dell’s US store on 18 Nov 2026. Memory prices have pushed every one of these machines up during 2026, so check the price on the day you buy.

Decode: the Mac’s lead

Decode is the speed at which a model writes its answer, in tokens per second. Each new token reads the model’s active weights once, so bandwidth sets the ceiling. This table shows plain decode at short context, with no speculative decoding:

ModelMac Studio M5 MaxMac Studio M5 UltraDGX Spark (not in the index)Strix Halo (not in the index)
gpt-oss-120b (MXFP4)community92 tok/ssourced130 tok/s58.7 tok/s (llama.cpp b7941, CUDA)56.6 tok/s (llama.cpp, Vulkan RADV); 51.8 on ROCm 7.2.3
gpt-oss-20b (MXFP4)community141 tok/scommunity184 tok/s83.4 tok/s (llama.cpp b7941)79.8 tok/s (Vulkan RADV)
Qwen 3.6-35B-A3B (4-bit)community130 tok/scommunity153 tok/sNo accountable figure found60.4 tok/s (UD-Q4_K_XL, Vulkan RADV)
Llama 70B class (4-bit)sourced13 tok/ssourced25 tok/s4.42 tok/s (Llama 3.1 70B q4_K_M, Ollama 0.12.6)About 5.0 tok/s (Shisa V2 70B, a Llama 3 70B fine-tune, i1-Q4_K_M, Vulkan, community)

Sources for the non-Mac figures:

The pattern holds wherever all three can run a model. On gpt-oss the DGX Spark and Strix Halo land within about 5% of each other, matching their near-identical bandwidth. An M5 Max Mac with twice that bandwidth lands well above both. NVIDIA’s own blog (24 Oct 2025) puts the Spark at 55.4 tok/s on gpt-oss-120b with llama.cpp, close to Gerganov’s figure. A cheaper Mac also gets into the same range: the 64 GB Mac mini M5 Pro ($2,699) holds gpt-oss-120b with about 2 GB to spare, and the index puts it at estimated~54 tok/s.

Speed falls as the context fills. At 32K tokens of context, Gerganov’s Spark run decodes gpt-oss-120b at 42.8 tok/s and Capitella’s Strix Halo run at 43.0 on Vulkan RADV (36.3 on ROCm). The index records Mac figures at the shortest context only, so it has no like-for-like Mac number yet.

Prompts and agents: the Spark’s lead

Prompt processing, or prefill, is how fast a machine reads the text you send before it starts answering. It is limited by GPU compute, not bandwidth, and here the ranking flips:

The LLMCheck index does not record prefill yet, so this page has no Mac prefill figure.

That matters most for coding agents, which send tens of thousands of tokens of context with every step. Artificial Analysis’s AA-AgentPerf-Local benchmark (29 Sep 2026) replays real agent sessions. It runs every machine with speculative decoding, so it is a different workload from the decode table above. On its laptops and workstations board, the DGX Spark finished fastest of the unified-memory machines; an RTX 5090 desktop on the same board was faster on every model that fits its 32 GB:

Model (AA-AgentPerf-Local, minutes)MacBook Pro M5 Pro, 64 GBAMD Ryzen AI Halo, 128 GBDGX Spark, 128 GB
Qwen3.8 27B37.634.517.3
Qwen3.6-35B-A3B11.911.77.3
Ling 3.0 Flashdoes not fit25.114.9

The only Mac on that board is an M5 Pro, the 307 GB/s chip. In its M5 Ultra review (21 Sep 2026), Tom’s Hardware found that Mac’s prompt speed beat the Spark on Qwen 3.8-27B, with decode “almost four times higher”. In a follow-up (10 Nov 2026), a 256 GB, 80-core M5 Ultra Mac Studio finished the AgentPerf-Local Qwen 3.8-27B run in just under half the time of one Dell GB10. Two clustered GB10s won on prompt speed for the larger DeepSeek V4.1 Flash and GLM 5.3 Flash. Those results are published mostly as charts, so this page quotes no numbers from them.

Software

Models above 128 GB

All three top out at 128 GB in one box. The Mac goes further:

To compare two Macs at these sizes, see Mac Studio M5 Max 128 GB vs M5 Ultra 256 GB.

Which to buy

Not sure which Mac? The Mac Advisor turns a budget and a use case into an exact configuration. The Mac-vs-Mac pages put any two configurations side by side.

Frequently Asked Questions

Is a Mac faster than an NVIDIA DGX Spark for local LLMs?

For generating text, an M5 Max or M5 Ultra Mac is. It reads memory at 614 or 1,228 GB/s against the Spark's 273 GB/s, and the LLMCheck index's M5 Max figure for gpt-oss-120b is well above the 58.7 tok/s published for the Spark with llama.cpp. For reading long prompts the Spark is far faster than Strix Halo, and in agent benchmarks it beats the M5 Pro MacBook Pro. The index does not yet record Mac prompt speeds.

Is AMD Strix Halo good for running local LLMs?

It is the cheapest way to get 128 GB of GPU-addressable memory, and it generates text about as fast as a DGX Spark: 56.6 tok/s on gpt-oss-120b with llama.cpp on Vulkan, according to Donato Capitella's published runs. Its weak spots are prompt processing, which is a quarter to a third of the Spark's on an empty context, and a software stack where Vulkan is still steadier than ROCm.

How much does a DGX Spark cost in November 2026?

NVIDIA launched the 128 GB Founders Edition at $3,999 in October 2025 and raised it to $4,699 in February 2026. The Register reported $6,950 on 2 October 2026, though NVIDIA's own pages showed no price on 18 November. A 64 GB version from OEMs starts at $4,999. Dell's GB10 workstation with 128 GB and 2 TB was $7,901 on Dell's US store on 18 November 2026.

What is the cheapest machine that runs gpt-oss-120b?

A Mac mini M5 Pro with 64 GB, at $2,699 from Apple: it holds gpt-oss-120b with about 2 GB to spare, which leaves little room for context. A 128 GB Strix Halo mini PC is the cheapest 128 GB machine, but no longer under $3,000: the Framework Desktop rose from $2,459 in January 2026 to $3,449 by September and was out of stock on 18 November 2026.

Can a DGX Spark or a Strix Halo box run models bigger than 128 GB?

Not on its own. Two DGX Sparks connect over one QSFP cable for 256 GB, and four need a 200 GbE switch. Strix Halo stops at 128 GB; AMD's Ryzen AI Max+ PRO 495 successor goes to 192 GB. A single Mac Studio M5 Ultra is sold with 256 or 512 GB.

Does LLMCheck benchmark the DGX Spark or Strix Halo?

No. LLMCheck runs no tests of its own, and its index covers Apple Silicon Macs only. The DGX Spark and Strix Halo figures on this page are quoted from the published sources linked beside them, with the runtime and date, and they never enter the index or its rankings.