The best local LLM for a Mac Studio M2 Ultra (192 GB) is MiMo-V2.6-Flash at ~54 tok/s (estimated). Its 192 GB fits 100 of the 130 models in the LLMCheck index, up to 428B-class. For everyday chat and coding, MiMo-V2.6-Flash is the sweet spot.
Considering this Mac?
Mac Studio M5 Ultra (256GB): build-to-order from Apple · Mac Studio M2 Ultra on Amazon Renewed → · compare every Mac for local AI →
Ranked by LLMCheck suitability (capability balanced against speed on the M2 Ultra); 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 | MiMo-V2.6-Flash raise GPU limit | 309B MoE | MIT | estimated54 tok/s | agent-ready | 49/50 |
| 2 | Qwen3.8-Flash-Next | 125B MoE | Qwen Custom | estimated57 tok/s | agent-ready | 47/50 |
| 3 | MiniMax-M3 raise GPU limit | 428B MoE | MiniMax Community | estimated42 tok/s | agent-ready | 48/50 |
| 4 | MiniMax-M3.1-Flash over GPU limit | 428B MoE | MiniMax Community | estimated25 tok/s | 48/50 | |
| 5 | Nemotron 3.5 Lightning | 30B MoE | OpenMDW | estimated150 tok/s | agent-ready | 36/50 |
| 6 | North Mini Code 1.0 | 30.5B MoE | Apache 2.0 | estimated132 tok/s | agent-ready | 36/50 |
| 7 | GLM-5.3-Flash | 320B MoE | MIT | estimated24 tok/s | agent-ready | 48/50 |
| 8 | KAT-Coder-V2.5 | 35B MoE | Apache 2.0 | estimated99 tok/s | agent-ready | 39/50 |
| 9 | DeepSeek V4 Flash Vision-Exp | 305B MoE | MIT | estimated32 tok/s | agent-ready | 47/50 |
| 10 | DeepSeek V4 Flash | 284B MoE | MIT | estimated31 tok/s | agent-ready | 47/50 |
| 11 | Qwen 3.6-35B-A3B | 35B MoE | Apache 2.0 | estimated105 tok/s | agent-ready | 38/50 |
| 12 | Qwen3.8-27B | 27.8B | Apache 2.0 | estimated38 tok/s | agent-ready | 46/50 |
Showing the top 12 of 100 models that fit in 192 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 176 GB this Mac’s GPU can use with the limit raised (87 of the 92 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 144 GB of this Mac’s 192 GB by default on macOS 15 and earlier, and about 161.3 GB on macOS 26 and later. Of the 100 models that fit, 6 need more than that at a short prompt: 4 need a raised limit and are tagged raise GPU limit; 2 need more than even the raised limit and are tagged over GPU limit. sudo sysctl iogpu.wired_limit_mb=180224 raises it to 176 GB, leaving 16 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: MiMo-V2.6-Flash needs about 170 GB, more than macOS lets the GPU use on this Mac by default (about 144 GB). Raise the limit first; it lasts until the Mac restarts:
The index has no verified Ollama tag for MiMo-V2.6-Flash, so start from its listing and copy the exact name there: Hugging Face repo → · model card →
GGUF builds run in LM Studio or llama.cpp; MLX builds run in LM Studio or mlx-lm.
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 MiMo-V2.6-Flash on M2 Ultra page for its figures and setup.
The models a 192 GB Mac can hold are big files: MiniMax-M3.1-Flash, the largest that fits here, is about a 187 GB download at the build it needs, and a few models that size fill an internal drive. A fast external Thunderbolt 5 or USB4 SSD keeps a model library within reach: Thunderbolt 5 external SSDs on Amazon →
Amazon affiliate link: as an Amazon Associate, LLMCheck earns from qualifying purchases.
Twice as fast: no Mac in the index runs MiMo-V2.6-Flash at twice this Mac’s estimated~54 tok/s; the fastest figure is the Mac Studio M5 Ultra (256 GB) at community77 tok/s, $9,499 new (Apple list price, checked 2026-11-14): build-to-order from Apple (Amazon does not stock this configuration).
For less: no cheaper Mac in the index runs MiMo-V2.6-Flash at 15+ tok/s.
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.
The Mac Studio M2 Ultra (192 GB) tops out at MiniMax-M3. A current Mac Studio M5 Ultra (256 GB) fits 8 more of the index’s models and reads memory at 1228 GB/s instead of 800 GB/s, so models also generate faster. The Mac Studio M5 Ultra (256 GB) is build-to-order from Apple; Amazon does not stock that configuration.
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 Studio M2 Ultra (192 GB). llmcheck.net. https://llmcheck.net/best-llm/mac-studio-m2-ultra-192gb/ (updated ). CC BY 4.0.
MiMo-V2.6-Flash (309B MoE, MIT) is the best all-round pick at ~54 tok/s (estimated) on the Mac Studio M2 Ultra. If you want maximum speed, LFM3-1.7B hits ~330 tok/s (estimated); for maximum capability, MiniMax-M3 still fits in 192 GB.
About 100 of the 130 models in the LLMCheck index fit in 192 GB of unified memory, from compact models up to MiniMax-M3 (428B MoE).
Yes. A dense 70B such as Llama 3.3 70B needs about 44 GB at 4-bit, which fits in 192 GB with room for context; the index puts it at ~15 tok/s (estimated) on this Mac.
192 GB is enough for dense models up to about 128B at 4-bit (the largest in the index that fits is Mistral Medium 3.5), and for mixture-of-experts models such as MiMo-V2.6-Flash (~54 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.