The best local LLM for a Mac Studio M5 Ultra (96 GB) is Qwen3.8-27B at 57 tok/s. With 96 GB of unified memory it runs 56 of the models we benchmark — from compact options up to 276B-class models. For everyday chat and coding, Qwen3.8-27B is the sweet spot. Full ranking below.
Ranked by LLMCheck suitability (capability balanced against speed on the M5 Ultra). Click a model for its full benchmark and setup. All speeds are index estimates (memory-bandwidth model, cross-referenced with sourced benchmarks where available) — submit a real run →
| # | Model | Size | License | Speed | Capability |
|---|---|---|---|---|---|
| 1 | Qwen3.8-27B | 27.8B | Apache 2.0 | 57 tok/s | 46/50 |
| 2 | KAT-Coder-V2.5 | 35B | Apache 2.0 | 106 tok/s est. | 39/50 |
| 3 | Inkling-Small | 276B | Apache 2.0 | 45 tok/s | 46/50 |
| 4 | Qwen 3.6-27B | 27B | Apache 2.0 | 61 tok/s est. | 44/50 |
| 5 | Nemotron 3.5 Lightning | 30B | OpenMDW | 123 tok/s est. | 36/50 |
| 6 | Qwen 3.6-35B-A3B | 35B | Apache 2.0 | 106 tok/s est. | 38/50 |
| 7 | Laguna S 2.1 | 118B | OpenMDW | 53 tok/s | 43/50 |
| 8 | Muse Glimmer 30B | 30B | Apache 2.0 | 55 tok/s est. | 42/50 |
| 9 | Gemma 4 26B-A4B | 26B | Apache 2.0 | 98 tok/s est. | 35/50 |
| 10 | Laguna XS 2.1 | 33B | OpenMDW | 113 tok/s est. | 33/50 |
| 11 | Gemma 4 31B | 31B | Apache 2.0 | 53 tok/s est. | 40/50 |
| 12 | Mistral Small 4 | 119B | Apache 2.0 | 86 tok/s | 34/50 |
Showing the top 12 of 56 models that fit in 96 GB. See the full leaderboard or all benchmarks.
The fastest way to get started is Ollama. Install it, then pull the top pick for your Mac:
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 Qwen3.8-27B on M5 Ultra benchmark page for exact settings.
The Mac Studio M5 Ultra (96 GB) comfortably runs 56 of the models we benchmark, led by Qwen3.8-27B. Grab one and start running LLMs offline today:
As an Amazon Associate, LLMCheck earns from qualifying purchases. Affiliate links cost you nothing extra and never influence our rankings.
Qwen3.8-27B (27.8B, Apache 2.0) is the best all-round pick at 57 tok/s on the M5 Ultra. If you want maximum speed, Maple Preview 20B-A1B hits 575 tok/s; for maximum capability, Inkling-Small still fits in 96 GB.
About 56 of the 81 models in the LLMCheck leaderboard fit in 96 GB of unified memory, from compact models up to Inkling-Small (276B).
Yes. A 70B model in Q4 quantization needs roughly 40–44 GB of memory, which fits in 96 GB with headroom for context.
96 GB is plenty for local AI — you can run capable 30B–70B-class models. Because Apple Silicon uses unified memory, that figure is both your system RAM and your VRAM.