Can I run GPT-oss 120B on a Mac?

Yes: GPT-oss 120B runs on 12 of 54 Macs in the LLMCheck index. It needs about 62 GB at a 4K context, 65 GB at 64K. Smallest current Mac: the Mac Studio M5 Ultra 96 GB; cheapest at 15+ tok/s: the Mac Studio M5 Max 128 GB ($5,099; 92 tok/s community-measured). Best measured: 142 tok/s on an M5 Ultra (published).

RAM at 4K
62
GB
RAM at 64K
65
GB, 16-bit KV
Macs that run it
12
of 54 configs
Best measured
sourced142
tok/s · M5 Ultra

GPT-oss 120B on every Mac

One row per Mac configuration, grouped by product line. The build is the one each speed is for; the context column is the longest prompt that fits inside macOS’s default GPU memory limit with a 16-bit KV cache (“raised” where the limit has to be raised first; “full” when the whole window fits). Speeds are decode tok/s from the LLMCheck index, each with its provenance mark. For large mixture-of-experts models (over 40B total), the formula’s typical error is about 34% (the median gap to 70 measured plain-decode runs; it tends to undershoot (median −33%), and an estimate scaled from the model’s own measured run is usually closer; method).

Runs fits macOS’s default GPU limit, 15+ tok/sRuns, slowly fits, under 15 tok/sRaise GPU limit fits once the limit is raised (command below)Over GPU limit fits RAM on paper, but even a safe raise is shortDoesn't fit needs more memory than the Mac has
MacVerdictBuildSpeedMax contextModel page
MacBook Air
MacBook Air M5 16 GBDoesn't fit————
MacBook Air M5 24 GBDoesn't fit————
MacBook Air M5 32 GBDoesn't fit————
MacBook Air M1 8 GB refurbDoesn't fit————
MacBook Air M4 16 GB usedDoesn't fit————
MacBook Air M2 16 GB refurbDoesn't fit————
MacBook Air M3 16 GB refurbDoesn't fit————
MacBook Air M1 16 GB refurbDoesn't fit————
MacBook Air M4 24 GB usedDoesn't fit————
MacBook Air M2 24 GB refurbDoesn't fit————
MacBook Air M3 24 GB refurbDoesn't fit————
MacBook Pro
MacBook Pro M6 16 GBDoesn't fit————
MacBook Pro M5 Pro 24 GBDoesn't fit————
MacBook Pro M6 24 GBDoesn't fit————
MacBook Pro M6 32 GBDoesn't fit————
MacBook Pro M5 Pro 48 GBDoesn't fit————
MacBook Pro M5 Max 64 GBOver GPU limit why ↓4.3 bpwestimated92 tok/s—M5 Max page
MacBook Pro M5 Pro 64 GBOver GPU limit why ↓4.3 bpwestimated54 tok/s——
MacBook Pro M5 Max 128 GBRunsMXFP4community92 tok/s128K (full)M5 Max page
MacBook Pro M3 Max 64 GB refurbOver GPU limit why ↓4.3 bpwestimated66 tok/s——
MacBook Pro M3 Max 128 GB refurbRunsMXFP4estimated65 tok/s128K (full)—
Mac mini
Mac mini M6 16 GBDoesn't fit————
Mac mini M5 Pro 24 GBDoesn't fit————
Mac mini M6 24 GBDoesn't fit————
Mac mini M6 32 GBDoesn't fit————
Mac mini M5 Pro 48 GBDoesn't fit————
Mac mini M5 Pro 64 GBOver GPU limit why ↓4.3 bpwestimated54 tok/s——
Mac mini M2 Pro 16 GB refurbDoesn't fit————
Mac mini M4 16 GB usedDoesn't fit————
Mac mini M4 Pro 24 GB usedDoesn't fit————
Mac mini M4 24 GB usedDoesn't fit————
Mac mini M2 Pro 32 GB refurbDoesn't fit————
Mac mini M4 Pro 48 GB usedDoesn't fit————
Mac mini M4 Pro 64 GB usedOver GPU limit why ↓4.3 bpwestimated48 tok/s——
iMac
iMac M6 16 GBDoesn't fit————
iMac M6 24 GBDoesn't fit————
iMac M6 32 GBDoesn't fit————
Mac Studio
Mac Studio M5 Max 36 GBDoesn't fit————
Mac Studio M5 Max 48 GBDoesn't fit————
Mac Studio M5 Max 64 GBOver GPU limit why ↓4.3 bpwestimated92 tok/s—M5 Max page
Mac Studio M5 Ultra 96 GBRunsMXFP4sourced130 tok/s128K (full)M5 Ultra page
Mac Studio M5 Max 128 GBRunsMXFP4community92 tok/s128K (full)M5 Max page
Mac Studio M5 Ultra 256 GBRunsMXFP4sourced130 tok/s128K (full)M5 Ultra page
Mac Studio M5 Ultra 512 GBRunsMXFP4sourced130 tok/s128K (full)M5 Ultra page
Mac Studio M2 Max 32 GB refurbDoesn't fit————
Mac Studio M1 Ultra 64 GB refurbOver GPU limit why ↓MXFP4estimated113 tok/s——
Mac Studio M4 Max 64 GB usedOver GPU limit why ↓4.3 bpwestimated87 tok/s——
Mac Studio M2 Max 64 GB refurbOver GPU limit why ↓4.3 bpwestimated66 tok/s——
Mac Studio M1 Ultra 128 GB refurbRunsMXFP4estimated113 tok/s128K (full)—
Mac Studio M2 Ultra 128 GB refurbRunsMXFP4estimated113 tok/s128K (full)—
Mac Studio M4 Max 128 GB usedRunsMXFP4estimated84 tok/s128K (full)—
Mac Studio M2 Ultra 192 GB refurbRunsMXFP4estimated113 tok/s128K (full)—
Mac Studio M3 Ultra 256 GB usedRunsMXFP4estimated115 tok/s128K (full)—
Mac Studio M3 Ultra 512 GB refurbRunsMXFP4estimated115 tok/s128K (full)—
Provenance key
sourced
Published figure — vendor or press review
estimated
community
Linked individual run — blog, forum or submission

Context memory is computed from its config.json (18 full layers + 18 sliding-window layers (window 128)): about 2.4 GB of 16-bit KV cache at 64K tokens. A q8 cache (llama.cpp -ctk q8_0 -ctv q8_0, Ollama OLLAMA_KV_CACHE_TYPE=q8_0) halves that part. Runtimes also grow a compute buffer with context, so a fit within about 1 GB of the limit is optimistic.

Raising the GPU memory limit

macOS lets the GPU wire only part of unified memory by default (Metal’s recommended working set; M5 and M6 Macs run macOS 26 or later, whose default is higher), and GPT-oss 120B needs about 62 GB. On the Macs marked Raise GPU limit it fits in RAM but not inside that default. The default is a performance threshold, not a hard wall: past it, the runtime moves part of it to the CPU or shortens the context (llama.cpp, Ollama), runs it slowly (MLX), or stops with an out-of-memory error. One command raises the limit until the next restart (sudo sysctl iogpu.wired_limit_mb=0 restores the default sooner); it leaves macOS a reserve, and other apps get less memory while the model is loaded. On the Macs marked Over GPU limit even that raise falls short: the model fits the Mac’s RAM on paper, but a smaller quant or a bigger Mac is the realistic route.

GPT-oss 120B: memory by quantization

Weights are 117B parameters at each format’s effective size (the index’s GB-per-billion table), plus about 2 GB for the runtime and a short context; longer contexts add the KV cache computed from the model’s config.json at 16 bits. It is a mixture-of-experts model: all 117B parameters sit in memory although only 5.1B are read per token. GPT-oss 120B ships natively in MXFP4, which the index costs at 0.54 GB per billion parameters; the other rows are third-party re-quantizations. The first row is the catalog’s own figure (about 62 GB at a 4K context), the one every verdict on this page uses. “Smallest Mac” is the smallest memory size among the Macs Apple sells new that holds it at a 4K context inside that Mac’s default GPU limit. Estimates, not measurements.

BuildWeightsRAM at 4Kat 32Kat 64KSmallest Mac
As catalogued—62 GB63 GB64 GB96 GB
Q2_K39.8 GB42 GB43 GB44 GB64 GB (48 GB with a raised limit)
Q3_K_M57.3 GB59 GB60 GB62 GB96 GB
Q4_K_M66.7 GB69 GB70 GB71 GB96 GB
Q5_K_M83.1 GB85 GB86 GB87 GB128 GB
Q6_K95.9 GB98 GB99 GB100 GB128 GB
Q8_0124.0 GB126 GB127 GB128 GB256 GB
F16234.0 GB236 GB237 GB238 GB512 GB (256 GB with a raised limit)

Measured runs of GPT-oss 120B

Every published (filled square) and community (hollow circle) run the index holds for GPT-oss 120B: 5 published, 2 community, each linked to its source. LLMCheck runs no tests of its own. One row is one measurement session; on each chip the table above shows the lower median of the eligible runs (plain decode at 5 bits or fewer), and a binned or “not an anchor” run is shown but never used as a reference.

ChipRuntimeQuantDecodeContextDateSource
M5 Ultra 256 GBllama.cppMXFP4sourced142 tok/sempty context, 128 generated2026-10-26vsesmart.cz
M5 Ultra 256 GBMLXMXFP4sourced130 tok/sempty context, 128 generated2026-10-26vsesmart.cz
M5 Ultra 256 GBMLXMXFP4community120 tok/s512-token random prompt, 128 generated, empty context, mean of 22026-10-15bohemiansilicon.com
M5 Max 128 GBllama.cppMXFP4sourced110 tok/srandom-token prompt, empty context2026-11-02vsesmart.cz
M5 Max 128 GBMLXMXFP4sourced94 tok/srandom-token prompt, empty context2026-11-02vsesmart.cz
M5 Max 128 GBMLXMXFP4community92 tok/s1k2026-11-06omlx.ai
M5 Max 128 GBOllamaunstatedsourced75.65 tok/s—2026-09-21tbreak.com

Recommended setup

The quickest start is Ollama. Pull the model by its tag (checked on ollama.com, 2026-11-14; the file is about 65 GB):

ollama run gpt-oss:120b

Prefer a window to a terminal? LM Studio downloads the same model with one click. · model card →

On the Mac Studio M5 Ultra (96 GB), the smallest current Mac that runs it inside the default GPU limit, the index figure is 130 tok/s (published figure) at MXFP4; keep the context at or under 128K there, or use a q8 KV cache. The gpt-oss:120b file fits there.

Step-by-step: the Ollama install guide.

🛒 Which Mac to get for GPT-oss 120B

Mac Studio M5 Ultra (96 GB) — smallest current Mac that runs it, $5,499 new (Apple list price, checked 2026-11-14), 130 tok/s (published figure). Mac Studio M5 Max (128 GB) — cheapest at 15+ tok/s, $5,099 new (Apple list price, checked 2026-11-14), 92 tok/s (community run). Amazon does not stock the Mac Studio M5 Ultra (96 GB): Apple builds it to order. Amazon does not stock the Mac Studio M5 Max (128 GB): Apple builds it to order.

FAQ: GPT-oss 120B on a Mac

Can I run GPT-oss 120B on a Mac?

Yes. GPT-oss 120B needs about 62 GB of unified memory at a short (4K-token) context, so it runs on 12 of the 54 Mac configurations in the LLMCheck index. On 9 more it fits the RAM, but not even a safely raised GPU limit. The smallest Mac Apple sells new that runs it is the Mac Studio M5 Ultra (96 GB).

How much RAM does GPT-oss 120B need?

About 62 GB at a 4K context: the weights plus about 2 GB for the runtime. Context adds KV cache: about 64 GB in total at 32K and 65 GB at 64K with the default 16-bit cache (about 64 GB at 64K with a q8 cache). These are estimates from the model’s config.json, not measurements.

How fast does GPT-oss 120B run on a Mac?

On an M5 Max, the chip the leaderboard quotes it on, the index figure is 92 tok/s (community run) at MXFP4. The best measured figure is 142 tok/s on an M5 Ultra 256 GB (published figure, MXFP4). Every Mac’s figure is in the table on this page with its provenance: published, community run or estimated.

What is the cheapest Mac that runs GPT-oss 120B well?

Among Macs Apple sells new, the Mac Studio M5 Max (128 GB), $5,099 new (Apple list price, checked 2026-11-14), is the cheapest that reaches 15 tok/s with GPT-oss 120B, at 92 tok/s (community run).

Do I need to raise the GPU memory limit to run GPT-oss 120B?

On Mac mini M5 Pro 64 GB, Mac Studio M5 Max 64 GB, MacBook Pro M5 Max 64 GB and 6 more even a safely raised limit is short: expect heavy swapping, a CPU split or an out-of-memory error. From 96 GB up, every Mac Apple sells new holds it inside the default limit.

Which quantization of GPT-oss 120B should I use?

GPT-oss 120B ships natively in MXFP4 (about 63 GB of weights); re-quantizing it gains little. The index’s reference figure (92 tok/s on an M5 Max) is for its MXFP4 build.

How much context can GPT-oss 120B use on a Mac?

Its own window is 128K tokens. At macOS’s default GPU limit, the Mac Studio M5 Ultra (96 GB) holds about 128K (the full window).

Cite this page

LLMCheck. (2026). Can I run GPT-oss 120B on a Mac? llmcheck.net. https://llmcheck.net/can-i-run/gpt-oss-120b/ (updated ). CC BY 4.0.

How to cite LLMCheck →Data: CC BY 4.0Markdown version

Related

Figures from the LLMCheck index, data version 2026-11-18. Verdicts and context limits are estimates from each model’s catalog memory figure and config.json.