Yes: GPT-oss 20B runs on 53 of 54 Macs in the LLMCheck index. It needs about 13 GB at a 4K context, 15 GB at 64K. The smallest and cheapest current Mac at 15+ tok/s is the Mac mini M6 16 GB ($899; ~41 tok/s est.), with the GPU limit raised. Best measured: 184 tok/s on an M5 Ultra (community).
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 small mixture-of-experts models (40B total or less), the formula’s typical error is about 15% (the median gap to 88 measured plain-decode runs; it tends to undershoot (median −7%), and an estimate scaled from the model’s own measured run is usually closer; method).
| Mac | Verdict | Build | Speed | Max context | Model page |
|---|---|---|---|---|---|
| MacBook Air | |||||
| MacBook Air M5 16 GB | Raise GPU limit command ↓ | MXFP4 | estimated41 tok/s | — | — |
| MacBook Air M5 24 GB | Runs | MXFP4 | estimated41 tok/s | 128K (full) | — |
| MacBook Air M5 32 GB | Runs | MXFP4 | estimated41 tok/s | 128K (full) | — |
| MacBook Air M1 8 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Air M4 16 GB used | Raise GPU limit command ↓ | MXFP4 | estimated34 tok/s | — | M4 page |
| MacBook Air M2 16 GB refurb | Raise GPU limit command ↓ | MXFP4 | estimated29 tok/s | — | — |
| MacBook Air M3 16 GB refurb | Raise GPU limit command ↓ | MXFP4 | estimated29 tok/s | — | — |
| MacBook Air M1 16 GB refurb | Raise GPU limit command ↓ | MXFP4 | estimated21 tok/s | — | — |
| MacBook Air M4 24 GB used | Runs | MXFP4 | estimated34 tok/s | 123K | M4 page |
| MacBook Air M2 24 GB refurb | Runs | MXFP4 | estimated29 tok/s | 123K | — |
| MacBook Air M3 24 GB refurb | Runs | MXFP4 | estimated29 tok/s | 123K | — |
| MacBook Pro | |||||
| MacBook Pro M6 16 GB | Raise GPU limit command ↓ | MXFP4 | estimated41 tok/s | — | M6 page |
| MacBook Pro M5 Pro 24 GB | Runs | MXFP4 | estimated69 tok/s | 128K (full) | — |
| MacBook Pro M6 24 GB | Runs | unstated | community45 tok/s | 128K (full) | M6 page |
| MacBook Pro M6 32 GB | Runs | unstated | community45 tok/s | 128K (full) | M6 page |
| MacBook Pro M5 Pro 48 GB | Runs | MXFP4 | estimated69 tok/s | 128K (full) | — |
| MacBook Pro M5 Max 64 GB | Runs | MXFP4 | community141 tok/s | 128K (full) | M5 Max page |
| MacBook Pro M5 Pro 64 GB | Runs | MXFP4 | estimated69 tok/s | 128K (full) | — |
| MacBook Pro M5 Max 128 GB | Runs | MXFP4 | community141 tok/s | 128K (full) | M5 Max page |
| MacBook Pro M3 Max 64 GB refurb | Runs | MXFP4 | estimated121 tok/s | 128K (full) | — |
| MacBook Pro M3 Max 128 GB refurb | Runs | MXFP4 | estimated121 tok/s | 128K (full) | — |
| Mac mini | |||||
| Mac mini M6 16 GB | Raise GPU limit command ↓ | MXFP4 | estimated41 tok/s | — | M6 page |
| Mac mini M5 Pro 24 GB | Runs | MXFP4 | estimated69 tok/s | 128K (full) | — |
| Mac mini M6 24 GB | Runs | unstated | community45 tok/s | 128K (full) | M6 page |
| Mac mini M6 32 GB | Runs | unstated | community45 tok/s | 128K (full) | M6 page |
| Mac mini M5 Pro 48 GB | Runs | MXFP4 | estimated69 tok/s | 128K (full) | — |
| Mac mini M5 Pro 64 GB | Runs | MXFP4 | estimated69 tok/s | 128K (full) | — |
| Mac mini M2 Pro 16 GB refurb | Raise GPU limit command ↓ | MXFP4 | estimated51 tok/s | — | — |
| Mac mini M4 16 GB used | Raise GPU limit command ↓ | MXFP4 | estimated34 tok/s | — | M4 page |
| Mac mini M4 Pro 24 GB used | Runs | MXFP4 | estimated64 tok/s | 123K | M4 Pro page |
| Mac mini M4 24 GB used | Runs | MXFP4 | estimated34 tok/s | 123K | M4 page |
| Mac mini M2 Pro 32 GB refurb | Runs | MXFP4 | estimated51 tok/s | 128K (full) | — |
| Mac mini M4 Pro 48 GB used | Runs | MXFP4 | estimated64 tok/s | 128K (full) | M4 Pro page |
| Mac mini M4 Pro 64 GB used | Runs | MXFP4 | estimated64 tok/s | 128K (full) | M4 Pro page |
| iMac | |||||
| iMac M6 16 GB | Raise GPU limit command ↓ | MXFP4 | estimated41 tok/s | — | M6 page |
| iMac M6 24 GB | Runs | unstated | community45 tok/s | 128K (full) | M6 page |
| iMac M6 32 GB | Runs | unstated | community45 tok/s | 128K (full) | M6 page |
| Mac Studio | |||||
| Mac Studio M5 Max 36 GB | Runs | MXFP4 | estimated128 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Max 48 GB | Runs | MXFP4 | community141 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Max 64 GB | Runs | MXFP4 | community141 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Ultra 96 GB | Runs | MXFP4 | community184 tok/s | 128K (full) | M5 Ultra page |
| Mac Studio M5 Max 128 GB | Runs | MXFP4 | community141 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Ultra 256 GB | Runs | MXFP4 | community184 tok/s | 128K (full) | M5 Ultra page |
| Mac Studio M5 Ultra 512 GB | Runs | MXFP4 | community184 tok/s | 128K (full) | M5 Ultra page |
| Mac Studio M2 Max 32 GB refurb | Runs | MXFP4 | estimated121 tok/s | 128K (full) | — |
| Mac Studio M1 Ultra 64 GB refurb | Runs | MXFP4 | estimated151 tok/s | 128K (full) | — |
| Mac Studio M4 Max 64 GB used | Runs | MXFP4 | estimated135 tok/s | 128K (full) | M4 Max page |
| Mac Studio M2 Max 64 GB refurb | Runs | MXFP4 | estimated121 tok/s | 128K (full) | — |
| Mac Studio M1 Ultra 128 GB refurb | Runs | MXFP4 | estimated151 tok/s | 128K (full) | — |
| Mac Studio M2 Ultra 128 GB refurb | Runs | MXFP4 | estimated151 tok/s | 128K (full) | — |
| Mac Studio M4 Max 128 GB used | Runs | MXFP4 | estimated135 tok/s | 128K (full) | M4 Max page |
| Mac Studio M2 Ultra 192 GB refurb | Runs | MXFP4 | estimated151 tok/s | 128K (full) | — |
| Mac Studio M3 Ultra 256 GB used | Runs | MXFP4 | estimated152 tok/s | 128K (full) | — |
| Mac Studio M3 Ultra 512 GB refurb | Runs | MXFP4 | estimated152 tok/s | 128K (full) | — |
Context memory is computed from its config.json (12 full layers + 12 sliding-window layers (window 128)): about 1.6 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.
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 20B needs about 13 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.
sudo sysctl iogpu.wired_limit_mb=12288sudo sysctl iogpu.wired_limit_mb=12288Weights are 21B 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 21B parameters sit in memory although only 3.6B are read per token. GPT-oss 20B 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 13 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.
| Build | Weights | RAM at 4K | at 32K | at 64K | Smallest Mac |
|---|---|---|---|---|---|
| As catalogued | — | 13 GB | 14 GB | 15 GB | 24 GB |
| Q2_K | 7.1 GB | 9 GB | 10 GB | 11 GB | 16 GB |
| Q3_K_M | 10.3 GB | 12 GB | 13 GB | 14 GB | 24 GB |
| Q4_K_M | 12.0 GB | 14 GB | 15 GB | 15 GB | 24 GB |
| Q5_K_M | 14.9 GB | 17 GB | 18 GB | 18 GB | 24 GB |
| Q6_K | 17.2 GB | 19 GB | 20 GB | 21 GB | 32 GB |
| Q8_0 | 22.3 GB | 24 GB | 25 GB | 26 GB | 32 GB |
| F16 | 42.0 GB | 44 GB | 45 GB | 46 GB | 64 GB |
Every published (filled square) and community (hollow circle) run the index holds for GPT-oss 20B: 3 published, 4 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.
| Chip | Runtime | Quant | Decode | Context | Date | Source |
|---|---|---|---|---|---|---|
| M5 Ultra | MLX | MXFP4 | community184 tok/s | 512-token random prompt, 128 generated, empty context, mean of 2 | 2026-10-15 | bohemiansilicon.com |
| M5 Max | llama.cpp | MXFP4 | sourced156 tok/s | random-token prompt, empty context | 2026-11-02 | vsesmart.cz |
| M5 Max | MLX | MXFP4 | sourced142 tok/s | random-token prompt, empty context | 2026-11-02 | vsesmart.cz |
| M5 Max | MLX | MXFP4 | community140.5 tok/s | 1k | 2026-10-31 | omlx.ai |
| M5 Max | Ollama | unstated | sourced107.51 tok/s | — | 2026-09-21 | tbreak.com |
| M6 | MLX | 4-bit | community65.6 tok/s | 1k | 2026-11-14 | omlx.ai |
| M6 | LM Studio | unstated | community45 tok/s | — | 2026-10 | github.com |
The quickest start is Ollama. Pull the model by its tag (checked on ollama.com, 2026-11-14; the file is about 14 GB):
Prefer a window to a terminal? LM Studio downloads the same model with one click. · model card →
On the Mac mini M6 (24 GB), the smallest current Mac that runs it inside the default GPU limit, the index figure is 45 tok/s (community run) at unstated; keep the context at or under 128K there, or use a q8 KV cache. The gpt-oss:20b file fits there.
Step-by-step: the Ollama install guide.
Mac mini M6 (16 GB) — smallest current Mac that runs it and cheapest at 15+ tok/s, $899 new (Apple list price, checked 2026-11-14), ~41 tok/s (estimated) once the GPU memory limit is raised. For Ollama: the gpt-oss:20b tag is about 14 GB, so it needs a 24 GB Mac; on 16 GB use a ~4-bit build after raising the GPU limit.
As an Amazon Associate, LLMCheck earns from qualifying purchases. Affiliate links cost you nothing extra and never influence the verdicts.
Yes. GPT-oss 20B needs about 13 GB of unified memory at a short (4K-token) context, so it runs on 53 of the 54 Mac configurations in the LLMCheck index; 43 of them hold it inside macOS’s default GPU memory limit and 10 only after raising that limit. The smallest Mac Apple sells new that runs it is the Mac mini M6 (16 GB).
About 13 GB at a 4K context: the weights plus about 2 GB for the runtime. Context adds KV cache: about 14 GB in total at 32K and 15 GB at 64K with the default 16-bit cache (about 14 GB at 64K with a q8 cache). These are estimates from the model’s config.json, not measurements.
On an M5 Max, the chip the leaderboard quotes it on, the index figure is 141 tok/s (community run) at MXFP4. The best measured figure is 184 tok/s on an M5 Ultra 256 GB (community run, MXFP4). Every Mac’s figure is in the table on this page with its provenance: published, community run or estimated.
Among Macs Apple sells new, the Mac mini M6 (16 GB), $899 new (Apple list price, checked 2026-11-14), is the cheapest that reaches 15 tok/s with GPT-oss 20B, at ~41 tok/s (estimated), after raising the GPU memory limit.
On 10 of the 54 configurations, yes: iMac M6 16 GB, Mac mini M6 16 GB, MacBook Air M5 16 GB and 7 more. macOS lets the GPU use only part of unified memory by default (about 12 GB on the iMac M6 (16 GB)), and GPT-oss 20B needs about 13 GB. Running sudo sysctl iogpu.wired_limit_mb=12288 in Terminal raises the limit on a 16 GB Mac until the next restart. From 24 GB up, every Mac Apple sells new holds it inside the default limit.
GPT-oss 20B ships natively in MXFP4 (about 11 GB of weights); re-quantizing it gains little. The index’s reference figure (141 tok/s on an M5 Max) is for its MXFP4 build.
Its own window is 128K tokens. At macOS’s default GPU limit, the Mac mini M6 (24 GB) holds about 128K (the full window).
LLMCheck. (2026). Can I run GPT-oss 20B on a Mac? llmcheck.net. https://llmcheck.net/can-i-run/gpt-oss-20b/ (updated ). CC BY 4.0.