The best local LLM for a Mac Studio M3 Ultra (256 GB) is Qwen3.8-Flash-Next at 63 tok/s. With 256 GB of unified memory it runs 65 of the models we benchmark — from compact options up to 753B-class models. For everyday chat and coding, Qwen3.8-Flash-Next is the sweet spot. Full ranking below.
Considering this Mac?
Mac Studio M3 Ultra (256GB) on Amazon → · not sure? the Mac Advisor picks a config →
Ranked by LLMCheck suitability (capability balanced against speed on the M3 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-Flash-Next | 125B | Qwen Custom | 63 tok/s | 47/50 |
| 2 | GLM 5.2 | 753B | MIT | 15 tok/s | 50/50 |
| 3 | GLM-5.3 | 753B | GLM-5.3 | 12 tok/s | 50/50 |
| 4 | GLM-5.3-Flash | 320B | MIT | 25 tok/s | 48/50 |
| 5 | DeepSeek V4 Flash Vision-Exp | 305B | MIT | 33 tok/s | 47/50 |
| 6 | Qwen3.8-27B | 27.8B | Apache 2.0 | 40 tok/s est. | 46/50 |
| 7 | DeepSeek V4 Flash | 284B | MIT | 30 tok/s | 47/50 |
| 8 | Inkling-Small | 276B | Apache 2.0 | 30 tok/s | 46/50 |
| 9 | Qwen 3.6-27B | 27B | Apache 2.0 | 41 tok/s est. | 44/50 |
| 10 | Qwen3-235B-A22B | 235B | Apache 2.0 | 19 tok/s | 46/50 |
| 11 | KAT-Coder-V2.5 | 35B | Apache 2.0 | 71 tok/s est. | 39/50 |
| 12 | Laguna S 2.1 | 118B | OpenMDW | 35 tok/s est. | 43/50 |
Showing the top 12 of 65 models that fit in 256 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-Flash-Next on M3 Ultra benchmark page for exact settings.
The Mac Studio M3 Ultra (256 GB) comfortably runs 65 of the models we benchmark, led by Qwen3.8-Flash-Next. 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-Flash-Next (125B, Qwen Custom) is the best all-round pick at 63 tok/s on the M3 Ultra. If you want maximum speed, Maple Preview 20B-A1B hits 384 tok/s; for maximum capability, GLM 5.2 still fits in 256 GB.
About 65 of the 85 models in the LLMCheck leaderboard fit in 256 GB of unified memory, from compact models up to GLM 5.2 (753B).
Yes. A 70B model in Q4 quantization needs roughly 40–44 GB of memory, which fits in 256 GB with headroom for context.
256 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.