Yes: LFM2.5-24B-A2B runs on 43 of 54 Macs in the LLMCheck index. It needs about 17 GB at a 4K context, 19 GB at 64K. The smallest and cheapest current Mac that runs it at 15+ tok/s is the Mac mini M6 24 GB ($1,099; ~79 tok/s est.). Best measured figure: 184 tok/s on an M5 Max (published figure, Q4_K_M).
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 | Doesn't fit | — | — | — | — |
| MacBook Air M5 24 GB | Runs | 4-bit | community74 tok/s | 40K | M5 page |
| MacBook Air M5 32 GB | Runs | 4-bit | community74 tok/s | 128K (full) | M5 page |
| 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 | Raise GPU limit (macOS 15) command ↓ | 4-bit | community59 tok/s | M4 page | |
| MacBook Air M2 24 GB refurb | Raise GPU limit (macOS 15) command ↓ | Q4_K_M | estimated52 tok/s | — | |
| MacBook Air M3 24 GB refurb | Raise GPU limit (macOS 15) command ↓ | Q4_K_M | estimated52 tok/s | — | |
| MacBook Pro | |||||
| MacBook Pro M6 16 GB | Doesn't fit | — | — | — | — |
| MacBook Pro M5 Pro 24 GB | Runs | oQ4e | community114 tok/s | 40K | M5 Pro page |
| MacBook Pro M6 24 GB | Runs | Q4_K_M | estimated79 tok/s | 40K | — |
| MacBook Pro M6 32 GB | Runs | Q4_K_M | estimated79 tok/s | 128K (full) | — |
| MacBook Pro M5 Pro 48 GB | Runs | oQ4e | community114 tok/s | 128K (full) | M5 Pro page |
| MacBook Pro M5 Max 64 GB | Runs | Q4_K_M | sourced184 tok/s | 128K (full) | M5 Max page |
| MacBook Pro M5 Pro 64 GB | Runs | oQ4e | community114 tok/s | 128K (full) | M5 Pro page |
| MacBook Pro M5 Max 128 GB | Runs | Q4_K_M | sourced184 tok/s | 128K (full) | M5 Max page |
| MacBook Pro M3 Max 64 GB refurb | Runs | Q4_K_M | estimated90 tok/s | 128K (full) | — |
| MacBook Pro M3 Max 128 GB refurb | Runs | Q4_K_M | estimated90 tok/s | 128K (full) | — |
| Mac mini | |||||
| Mac mini M6 16 GB | Doesn't fit | — | — | — | — |
| Mac mini M5 Pro 24 GB | Runs | oQ4e | community114 tok/s | 40K | M5 Pro page |
| Mac mini M6 24 GB | Runs | Q4_K_M | estimated79 tok/s | 40K | — |
| Mac mini M6 32 GB | Runs | Q4_K_M | estimated79 tok/s | 128K (full) | — |
| Mac mini M5 Pro 48 GB | Runs | oQ4e | community114 tok/s | 128K (full) | M5 Pro page |
| Mac mini M5 Pro 64 GB | Runs | oQ4e | community114 tok/s | 128K (full) | M5 Pro page |
| 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 | Raise GPU limit (macOS 15) command ↓ | Q4_K_M | estimated90 tok/s | — | |
| Mac mini M4 24 GB used | Raise GPU limit (macOS 15) command ↓ | 4-bit | community59 tok/s | M4 page | |
| Mac mini M2 Pro 32 GB refurb | Runs | Q4_K_M | estimated87 tok/s | 128K (full) | — |
| Mac mini M4 Pro 48 GB used | Runs | Q4_K_M | estimated90 tok/s | 128K (full) | — |
| Mac mini M4 Pro 64 GB used | Runs | Q4_K_M | estimated90 tok/s | 128K (full) | — |
| iMac | |||||
| iMac M6 16 GB | Doesn't fit | — | — | — | — |
| iMac M6 24 GB | Runs | Q4_K_M | estimated79 tok/s | 40K | — |
| iMac M6 32 GB | Runs | Q4_K_M | estimated79 tok/s | 128K (full) | — |
| Mac Studio | |||||
| Mac Studio M5 Max 36 GB | Runs | Q4_K_M | estimated114 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Max 48 GB | Runs | Q4_K_M | sourced184 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Max 64 GB | Runs | Q4_K_M | sourced184 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Ultra 96 GB | Runs | Q4_K_M | estimated184 tok/s | 128K (full) | — |
| Mac Studio M5 Max 128 GB | Runs | Q4_K_M | sourced184 tok/s | 128K (full) | M5 Max page |
| Mac Studio M5 Ultra 256 GB | Runs | Q4_K_M | estimated184 tok/s | 128K (full) | — |
| Mac Studio M5 Ultra 512 GB | Runs | Q4_K_M | estimated184 tok/s | 128K (full) | — |
| Mac Studio M2 Max 32 GB refurb | Runs | Q4_K_M | estimated90 tok/s | 128K (full) | — |
| Mac Studio M1 Ultra 64 GB refurb | Runs | Q4_K_M | estimated114 tok/s | 128K (full) | — |
| Mac Studio M4 Max 64 GB used | Runs | 5-bit | community90 tok/s | 128K (full) | M4 Max page |
| Mac Studio M2 Max 64 GB refurb | Runs | Q4_K_M | estimated90 tok/s | 128K (full) | — |
| Mac Studio M1 Ultra 128 GB refurb | Runs | Q4_K_M | estimated114 tok/s | 128K (full) | — |
| Mac Studio M2 Ultra 128 GB refurb | Runs | Q4_K_M | estimated114 tok/s | 128K (full) | — |
| Mac Studio M4 Max 128 GB used | Runs | 5-bit | community90 tok/s | 128K (full) | M4 Max page |
| Mac Studio M2 Ultra 192 GB refurb | Runs | Q4_K_M | estimated114 tok/s | 128K (full) | — |
| Mac Studio M3 Ultra 256 GB used | Runs | 4-bit | community116 tok/s | 128K (full) | M3 Ultra page |
| Mac Studio M3 Ultra 512 GB refurb | Runs | 4-bit | community116 tok/s | 128K (full) | M3 Ultra page |
Context memory is computed from its config.json (hybrid short-conv: attention on 10 of 40 layers; conv layers keep a constant state): about 1.3 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 LFM2.5-24B-A2B needs about 17 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=20480Weights are 24B 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 24B parameters sit in memory although only 2.45B are read per token. The first row is the catalog’s own figure (about 17 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 | — | 17 GB | 18 GB | 18 GB | 24 GB |
| Q2_K | 8.2 GB | 10 GB | 11 GB | 11 GB | 16 GB |
| Q3_K_M | 11.8 GB | 14 GB | 14 GB | 15 GB | 24 GB |
| Q4_K_M | 13.7 GB | 16 GB | 16 GB | 17 GB | 24 GB |
| Q5_K_M | 17.0 GB | 19 GB | 20 GB | 20 GB | 32 GB |
| Q6_K | 19.7 GB | 22 GB | 22 GB | 23 GB | 32 GB |
| Q8_0 | 25.4 GB | 27 GB | 28 GB | 29 GB | 36 GB |
| F16 | 48.0 GB | 50 GB | 51 GB | 51 GB | 64 GB |
Every published (filled square) and community (hollow circle) run the index holds for LFM2.5-24B-A2B: 1 published, 8 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 |
|---|---|---|---|---|---|---|
| M3 Ultra | MLX | 4-bit | community115.6 tok/s | 4k | 2026-10-21 | omlx.ai |
| M5 Max | llama.cpp | Q4_K_M | sourced184 tok/s | 4K-token input | 2026-10-13 | huggingface.co |
| M5 Max | MLX | 8-bit | community131.5 tok/s | 1k | 2026-10-20 | omlx.ai |
| M4 Max | MLX | 5-bit | community89.7 tok/s | 4k | 2026-10-15 | omlx.ai |
| M5 Pro | MLX | 4-bit | community122 tok/s | 4k | 2026-10-19 | omlx.ai |
| M5 Pro | MLX | oQ4e | community113.5 tok/s | 1k | 2026-10-19 | omlx.ai |
| M5 | MLX | 4-bit | community74.2 tok/s | 4k | 2026-10-17 | omlx.ai |
| M4 | MLX | 4-bit | community59.9 tok/s | 1k | 2026-10-16 | omlx.ai |
| M4 | MLX | 4-bit | community59.3 tok/s | 1k | 2026-10-22 | omlx.ai |
The quickest start is Ollama. Pull the model by its tag (checked on ollama.com, 2026-11-14; the file is about 15 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 ~79 tok/s (estimated) at Q4_K_M; keep the context at or under 40K there, or use a q8 KV cache. The lfm2.5:24b file fits there.
Step-by-step: the Ollama install guide.
Mac mini M6 (24 GB) — smallest current Mac that runs it and cheapest at 15+ tok/s, $1,099 new (Apple list price, checked 2026-11-14), ~79 tok/s (estimated).
As an Amazon Associate, LLMCheck earns from qualifying purchases. Affiliate links cost you nothing extra and never influence the verdicts.
Yes. LFM2.5-24B-A2B needs about 17 GB of unified memory at a short (4K-token) context, so it runs on 43 of the 54 Mac configurations in the LLMCheck index; 38 of them hold it inside macOS’s default GPU memory limit and 5 only after raising that limit. The smallest Mac Apple sells new that runs it is the Mac mini M6 (24 GB).
About 17 GB at a 4K context: the weights plus about 2 GB for the runtime. Context adds KV cache: about 18 GB in total at 32K and 19 GB at 64K with the default 16-bit cache (about 18 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 184 tok/s (published figure) at Q4_K_M. The best measured figure is 184 tok/s on an M5 Max (published figure, Q4_K_M). 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 (24 GB), $1,099 new (Apple list price, checked 2026-11-14), is the cheapest that reaches 15 tok/s with LFM2.5-24B-A2B, at ~79 tok/s (estimated).
On 5 of the 54 configurations, yes: Mac mini M4 24 GB (used), Mac mini M4 Pro 24 GB (used), MacBook Air M2 24 GB (used) and 2 more. macOS lets the GPU use only part of unified memory by default (about 16 GB on the Mac mini M4 (24 GB)), and LFM2.5-24B-A2B needs about 17 GB. Running sudo sysctl iogpu.wired_limit_mb=20480 in Terminal raises the limit on a 24 GB Mac until the next restart. From 24 GB up, every Mac Apple sells new holds it inside the default limit.
Q4_K_M (about 14 GB of weights) is the common 4-bit GGUF choice. Q8_0 (about 25 GB) is closer to full quality and needs a 36 GB Mac at the default GPU limit. Q3_K_M and Q2_K shrink it to about 12 and 8 GB at a quality cost. The index’s reference figure (184 tok/s on an M5 Max) is for its Q4_K_M build.
Its own window is 128K tokens. At macOS’s default GPU limit, the Mac mini M6 (24 GB) holds about 40K. The full window needs about 20 GB, which a 32 GB Mac holds at the default limit. A q8 KV cache roughly doubles these lengths.
LLMCheck. (2026). Can I run LFM2.5-24B-A2B on a Mac? llmcheck.net. https://llmcheck.net/can-i-run/lfm25-24b-a2b/ (updated ). CC BY 4.0.