The best local LLM for a Mac Studio M4 Ultra (96 GB) is Qwen3.8-27B at 49 tok/s (estimated). With 96 GB of unified memory it runs 57 of the models in the LLMCheck index — from compact options up to 276B-class models. For everyday chat and coding, Qwen3.8-27B is the sweet spot. Full ranking below.
Considering this Mac?
Mac Studio M4 Ultra (96GB) on Amazon → · not sure? the Mac Advisor picks a config →
Ranked by LLMCheck suitability (capability balanced against speed on the M4 Ultra). Click a model for its full benchmark and setup. Every speed carries its provenance mark: most are index estimates from the memory-bandwidth model, and a filled square is a published measurement — submit a real run →
| # | Model | Size | License | Speed | Capability |
|---|---|---|---|---|---|
| 1 | Qwen3.8-27B | 27.8B | Apache 2.0 | estimated49 tok/s | 46/50 |
| 2 | Inkling-Small | 276B | Apache 2.0 | estimated40 tok/s | 46/50 |
| 3 | Qwen 3.6-27B | 27B | Apache 2.0 | estimated52 tok/s | 44/50 |
| 4 | KAT-Coder-V2.5 | 35B | Apache 2.0 | estimated89 tok/s | 39/50 |
| 5 | Laguna S 2.1 | 118B | OpenMDW | estimated48 tok/s | 43/50 |
| 6 | Nemotron 3.5 Lightning | 30B | OpenMDW | estimated102 tok/s | 36/50 |
| 7 | Qwen 3.6-35B-A3B | 35B | Apache 2.0 | estimated84 tok/s | 38/50 |
| 8 | Muse Glimmer 30B | 30B | Apache 2.0 | estimated48 tok/s | 42/50 |
| 9 | Qwen 3.5 122B-A10B | 122B | Apache 2.0 | estimated71 tok/s | 38/50 |
| 10 | Gemma 4 26B-A4B | 26B | Apache 2.0 | estimated88 tok/s | 35/50 |
| 11 | Gemma 4 31B | 31B | Apache 2.0 | estimated46 tok/s | 40/50 |
| 12 | Laguna XS 2.1 | 33B | OpenMDW | estimated94 tok/s | 33/50 |
Showing the top 12 of 57 models that fit in 96 GB. See the full leaderboard or all benchmarks.
The fastest way to get started is Ollama. Install it, then find the top pick for your Mac in its library and copy the exact tag it lists:
Find Qwen3.8-27B in the Ollama library → · model card on Hugging Face →
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 M4 Ultra benchmark page for exact settings.
The Mac Studio M4 Ultra (96 GB) comfortably runs 57 of the models in the LLMCheck index, 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 49 tok/s (estimated) on the M4 Ultra. If you want maximum speed, Maple Preview 20B-A1B hits 503 tok/s (estimated); for maximum capability, Inkling-Small still fits in 96 GB.
About 57 of the 89 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.