Yes: Qwen3-235B-A22B runs on 5 of the 54 Macs in the LLMCheck index. It needs about 128 GB at a 4K context, 140 GB at 64K. The smallest and cheapest current Mac that runs it at 15+ tok/s is the Mac Studio M5 Ultra 256 GB ($9,499; 46 tok/s published). Best measured figure: 46 tok/s on an M5 Ultra (published).
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).
| Mac | Verdict | Build | Speed | Max context | Model page |
|---|---|---|---|---|---|
| MacBook Air | |||||
| MacBook Air M5 16 GB | Doesn't fit | — | — | — | — |
| MacBook Air M5 24 GB | Doesn't fit | — | — | — | — |
| MacBook Air M5 32 GB | Doesn't fit | — | — | — | — |
| MacBook Air M1 8 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Air M4 16 GB used | Doesn't fit | — | — | — | — |
| MacBook Air M2 16 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Air M3 16 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Air M1 16 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Air M4 24 GB used | Doesn't fit | — | — | — | — |
| MacBook Air M2 24 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Air M3 24 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Pro | |||||
| MacBook Pro M6 16 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M5 Pro 24 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M6 24 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M6 32 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M5 Pro 48 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M5 Max 64 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M5 Pro 64 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M5 Max 128 GB | Over GPU limit why ↓ | 4.5 bpw | estimated26 tok/s | — | — |
| MacBook Pro M3 Max 64 GB refurb | Doesn't fit | — | — | — | — |
| MacBook Pro M3 Max 128 GB refurb | Over GPU limit why ↓ | 4.5 bpw | estimated17 tok/s | — | — |
| Mac mini | |||||
| Mac mini M6 16 GB | Doesn't fit | — | — | — | — |
| Mac mini M5 Pro 24 GB | Doesn't fit | — | — | — | — |
| Mac mini M6 24 GB | Doesn't fit | — | — | — | — |
| Mac mini M6 32 GB | Doesn't fit | — | — | — | — |
| Mac mini M5 Pro 48 GB | Doesn't fit | — | — | — | — |
| Mac mini M5 Pro 64 GB | Doesn't fit | — | — | — | — |
| Mac mini M2 Pro 16 GB refurb | Doesn't fit | — | — | — | — |
| Mac mini M4 16 GB used | Doesn't fit | — | — | — | — |
| Mac mini M4 Pro 24 GB used | Doesn't fit | — | — | — | — |
| Mac mini M4 24 GB used | Doesn't fit | — | — | — | — |
| Mac mini M2 Pro 32 GB refurb | Doesn't fit | — | — | — | — |
| Mac mini M4 Pro 48 GB used | Doesn't fit | — | — | — | — |
| Mac mini M4 Pro 64 GB used | Doesn't fit | — | — | — | — |
| iMac | |||||
| iMac M6 16 GB | Doesn't fit | — | — | — | — |
| iMac M6 24 GB | Doesn't fit | — | — | — | — |
| iMac M6 32 GB | Doesn't fit | — | — | — | — |
| Mac Studio | |||||
| Mac Studio M5 Max 36 GB | Doesn't fit | — | — | — | — |
| Mac Studio M5 Max 48 GB | Doesn't fit | — | — | — | — |
| Mac Studio M5 Max 64 GB | Doesn't fit | — | — | — | — |
| Mac Studio M5 Ultra 96 GB | Doesn't fit | — | — | — | — |
| Mac Studio M5 Max 128 GB | Over GPU limit why ↓ | 4.5 bpw | estimated26 tok/s | — | — |
| Mac Studio M5 Ultra 256 GB | Runs | 4-bit | sourced46 tok/s | 128K (full) | M5 Ultra page |
| Mac Studio M5 Ultra 512 GB | Runs | 4-bit | sourced46 tok/s | 128K (full) | M5 Ultra page |
| Mac Studio M2 Max 32 GB refurb | Doesn't fit | — | — | — | — |
| Mac Studio M1 Ultra 64 GB refurb | Doesn't fit | — | — | — | — |
| Mac Studio M4 Max 64 GB used | Doesn't fit | — | — | — | — |
| Mac Studio M2 Max 64 GB refurb | Doesn't fit | — | — | — | — |
| Mac Studio M1 Ultra 128 GB refurb | Over GPU limit why ↓ | 4.5 bpw | estimated32 tok/s | — | — |
| Mac Studio M2 Ultra 128 GB refurb | Over GPU limit why ↓ | 4.5 bpw | estimated32 tok/s | — | — |
| Mac Studio M4 Max 128 GB used | Over GPU limit why ↓ | 4.5 bpw | estimated23 tok/s | — | — |
| Mac Studio M2 Ultra 192 GB refurb | Runs | Q4_K_M | estimated31 tok/s | 85K | — |
| Mac Studio M3 Ultra 256 GB used | Runs | Q4_K_M | estimated32 tok/s | 128K (full) | — |
| Mac Studio M3 Ultra 512 GB refurb | Runs | Q4_K_M | estimated32 tok/s | 128K (full) | — |
Context memory is computed from its config.json (GQA on all 94 layers (4 kv heads x 128)): about 12.6 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 Qwen3-235B-A22B needs about 128 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.
sudo sysctl iogpu.wired_limit_mb=114688, 112 GB) is short of the ~128 GB it needs: expect heavy swapping, a CPU split or an out-of-memory errorsudo sysctl iogpu.wired_limit_mb=114688, 112 GB) is short of the ~128 GB it needs: expect heavy swapping, a CPU split or an out-of-memory errorWeights are 235B 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 235B parameters sit in memory although only 22B are read per token. The first row is the catalog’s own figure (about 128 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 | — | 128 GB | 134 GB | 140 GB | 256 GB |
| Q2_K | 79.9 GB | 82 GB | 87 GB | 94 GB | 128 GB |
| Q3_K_M | 115.1 GB | 117 GB | 123 GB | 129 GB | 256 GB |
| Q4_K_M | 133.9 GB | 136 GB | 141 GB | 148 GB | 256 GB |
| Q5_K_M | 166.8 GB | 169 GB | 174 GB | 181 GB | 256 GB |
| Q6_K | 192.7 GB | 195 GB | 200 GB | 207 GB | 256 GB |
| Q8_0 | 249.1 GB | 251 GB | 257 GB | 263 GB | 512 GB |
| F16 | 470.0 GB | 472 GB | 478 GB | 484 GB | 512 GB |
Every published (filled square) and community (hollow circle) run the index holds for Qwen3-235B-A22B: 1 published, 0 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 | 4-bit | sourced46 tok/s | empty context, 128 generated | 2026-10-26 | vsesmart.cz |
The quickest start is Ollama. Pull the model by its tag (checked on ollama.com, 2026-11-14; the file is about 142 GB):
Prefer a window to a terminal? LM Studio downloads the same model with one click. · model card →
On the Mac Studio M5 Ultra (256 GB), the smallest current Mac that runs it inside the default GPU limit, the index figure is 46 tok/s (published figure) at 4-bit; keep the context at or under 128K there, or use a q8 KV cache. The qwen3:235b-a22b file fits there.
Step-by-step: the Ollama install guide.
Mac Studio M5 Ultra (256 GB) — smallest current Mac that runs it and cheapest at 15+ tok/s, $9,499 new (Apple list price, checked 2026-11-14), 46 tok/s (published figure). Amazon does not stock the Mac Studio M5 Ultra (256 GB): Apple builds it to order.
Yes. Qwen3-235B-A22B needs about 128 GB of unified memory at a short (4K-token) context, so it runs on 5 of the 54 Mac configurations in the LLMCheck index. On 6 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 (256 GB).
About 128 GB at a 4K context: the weights plus about 2 GB for the runtime. Context adds KV cache: about 134 GB in total at 32K and 140 GB at 64K with the default 16-bit cache (about 134 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 ~26 tok/s (estimated) at 4.5 bpw. The best measured figure is 46 tok/s on an M5 Ultra 256 GB (published figure, 4-bit). 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 Studio M5 Ultra (256 GB), $9,499 new (Apple list price, checked 2026-11-14), is the cheapest that reaches 15 tok/s with Qwen3-235B-A22B, at 46 tok/s (published figure).
On Mac Studio M5 Max 128 GB, MacBook Pro M5 Max 128 GB, Mac Studio M1 Ultra 128 GB (used) and 3 more even a safely raised limit is short: expect heavy swapping, a CPU split or an out-of-memory error. From 256 GB up, every Mac Apple sells new holds it inside the default limit.
Q4_K_M (about 134 GB of weights) is the common 4-bit GGUF choice. Q8_0 (about 249 GB) is closer to full quality and needs a 512 GB Mac at the default GPU limit. Q3_K_M and Q2_K shrink it to about 115 and 80 GB at a quality cost. The index’s reference figure (26 tok/s on an M5 Max) is for its 4.5 bpw build.
Its own window is 128K tokens. At macOS’s default GPU limit, the Mac Studio M5 Ultra (256 GB) holds about 128K (the full window).
LLMCheck. (2026). Can I run Qwen3-235B-A22B on a Mac? llmcheck.net. https://llmcheck.net/can-i-run/qwen3-235b-a22b/ (updated ). CC BY 4.0.