The best local LLM for a Mac mini M2 Pro (32 GB) is Qwen 3.6-35B-A3B at ~71 tok/s (estimated), the top-ranked model that reaches 15 tok/s here. Its 32 GB fits 63 of the 130 models in the LLMCheck index, up to 39.8B-class. The most capable, Qwen3.8-27B, runs at ~11 tok/s (estimated): slow for chat.
Considering this Mac?
Mac Studio M5 Max (36GB) on Amazon → · compare every Mac for local AI →
Ranked by LLMCheck suitability (capability balanced against speed on the M2 Pro); the top pick is the best-ranked model that reaches 15 tok/s here, has a standard Mac runtime and fits the GPU memory limit (raised if need be), and slower models are tagged slow. Click a model for its full figures and setup. Every speed carries its provenance mark: most are index estimates from the memory-bandwidth model, a filled square is a published figure and a hollow circle a community run — submit a real run →
| # | Model | Size | License | Speed | 64K context | Capability |
|---|---|---|---|---|---|---|
| 1 | Qwen 3.6-35B-A3B raise GPU limit (macOS 15) | 35B MoE | Apache 2.0 | estimated71 tok/s | agent-ready | 38/50 |
| 2 | Qwen3.8-27B slow | 27.8B | Apache 2.0 | estimated11 tok/s | agent-ready | 46/50 |
| 3 | Nanbeige4.3-33B slow | 33.1B | Apache 2.0 | estimated8 tok/s | 46/50 | |
| 4 | LFM2.5-24B-A2B | 24B MoE | LFM Open | estimated87 tok/s | agent-ready | 35/50 |
| 5 | KAT-Coder-V2.5 raise GPU limit (macOS 15) | 35B MoE | Apache 2.0 | estimated53 tok/s | agent-ready | 39/50 |
| 6 | Qwen 3.6-27B slow | 27B | Apache 2.0 | estimated11 tok/s | agent-ready | 44/50 |
| 7 | Recursive Flint | 39.8B MoE | Apache 2.0 | estimated68 tok/s | agent-ready | 37/50 |
| 8 | Muse Glimmer 30B slow | 30B | Apache 2.0 | estimated7 tok/s | agent-ready | 42/50 |
| 9 | Nemotron 3.5 Lightning | 30B MoE | OpenMDW | estimated51 tok/s | agent-ready | 36/50 |
| 10 | Gemma 4 26B-A4B | 26B MoE | Apache 2.0 | estimated59 tok/s | agent-ready | 35/50 |
| 11 | Raon-1 mini preview | 28.8B MoE | Apache 2.0 | estimated65 tok/s | agent-ready | 34/50 |
| 12 | GLM-4.7-Flash | 31B MoE | MIT | estimated55 tok/s | agent-ready | 35/50 |
Showing the top 12 of 63 models that fit in 32 GB. See the full leaderboard or all benchmarks.
Context: “fits” means the weights at a short prompt. Coding agents (Claude Code, OpenCode, Codex CLI) hold 64K tokens or more, and every token adds KV cache: agent-ready marks a model whose weights plus a 64K-token 16-bit KV cache still fit in the 28 GB this Mac’s GPU can use with the limit raised (51 of the 58 models here with a KV figure). That is an estimate from each model’s config.json, not a measurement; a q8 KV cache halves the cache part, and — means the index has no KV figure for the model yet. The GPU memory limit below applies on top.
GPU memory limit: macOS lets the GPU use about 21.3 GB of this Mac’s 32 GB by default on macOS 15 and earlier, and about 25 GB on macOS 26 and later. Of the 63 models that fit, 9 need more than that at a short prompt and are tagged raise GPU limit. sudo sysctl iogpu.wired_limit_mb=28672 raises it to 28 GB, leaving 4 GB for macOS, until the Mac restarts (sudo sysctl iogpu.wired_limit_mb=0 restores the default). How the limit is read →
GPU memory limit: Qwen 3.6-35B-A3B needs about 24 GB, more than macOS lets the GPU use on this Mac by default on macOS 15 and earlier (about 21.3 GB). Raise the limit first; it lasts until the Mac restarts:
The fastest way to get started is Ollama. Install it, then pull the model by its tag (checked on ollama.com, 2026-11-14):
ollama run qwen3.6:35b-a3b · model card →
Prefer a GUI? LM Studio gives you a one-click download and chat window. For step-by-step help see our Ollama install guide, or open the Qwen 3.6-35B-A3B on M2 Pro page for its figures and setup.
Twice as fast: the cheapest Mac in the index that runs Qwen 3.6-35B-A3B at 2× or more of this Mac’s estimated~71 tok/s is the Mac Studio M5 Ultra (96 GB) at community153 tok/s, $5,499 new (Apple list price, checked 2026-11-14): build-to-order from Apple (Amazon does not stock this configuration).
For less: the cheapest Mac that still runs Qwen 3.6-35B-A3B at 15+ tok/s is the MacBook Air M2 (24 GB) at estimated~43 tok/s, about $850 refurbished or used: search Amazon Renewed listings →.
Speeds are each Mac’s own figure from the index, with its provenance mark. Prices come from data/macs.json: Apple’s list price for a new Mac (dated), an approximate street price for a used or refurbished one. Amazon stock checked 2026-11-18. Amazon links are affiliate links: as an Amazon Associate, LLMCheck earns from qualifying purchases.
The Mac mini M2 Pro (32 GB) tops out at Recursive Flint. A current Mac Studio M5 Max (36 GB) fits 1 more of the index’s models and reads memory at 460 GB/s instead of 200 GB/s, so models also generate faster:
As an Amazon Associate, LLMCheck earns from qualifying purchases. Affiliate links cost you nothing extra and never influence our rankings.
LLMCheck. (2026). Best Local LLMs for the Mac mini M2 Pro (32 GB). llmcheck.net. https://llmcheck.net/best-llm/mac-mini-m2-pro-32gb/ (updated ). CC BY 4.0.
Qwen 3.6-35B-A3B (35B MoE, Apache 2.0) is the best all-round pick at ~71 tok/s (estimated) on the Mac mini M2 Pro. If you want maximum speed, Ling-3.0-tiny hits ~139 tok/s (estimated); Qwen3.8-27B is more capable but runs at only ~11 tok/s (estimated) here.
About 63 of the 130 models in the LLMCheck index fit in 32 GB of unified memory, from compact models up to Recursive Flint (39.8B MoE).
No. A dense 70B such as Llama 3.3 70B needs about 44 GB at 4-bit; with 32 GB, stick to models like Qwen 3.6-35B-A3B. For a 70B, look at a 64 GB+ Mac.
32 GB is enough for dense models up to about 33B at 4-bit (the largest in the index that fits is Nanbeige4.3-33B), and for mixture-of-experts models such as KAT-Coder-V2.5 (~53 tok/s (estimated)). As a rule of thumb a 4-bit model needs about 0.57 GB per billion parameters plus ~2 GB, and macOS needs its own share of memory too. Because Apple Silicon uses unified memory, the same pool serves as system RAM and GPU memory.