M3 Ultra (80-core GPU, 256 GB) — benchmark record
M3 Ultra (80-core GPU, 256 GB) local LLM benchmarks on Apple Silicon Mac with 3 published rows across 3 models. Peak published speed is 177.9 tok/s on Llama 3.2 1B. Published runtimes include llamafile.
Quick take
Best published speed here is 177.9 tok/s on Llama 3.2 1B at Q4_K - Medium. This page is evidence, not the full buying answer.
Based on 3 external benchmarks; no lab runs yet.
Published runtimes: llamafile.
Current published coverage
Published model coverage includes Llama 3.2 1B, Llama 3.1 8B, Qwen 2.5 14B. Published runtime coverage on this chip includes llamafile. Fastest published row is 177.9 tok/s on Llama 3.2 1B at Q4_K - Medium.
Other published M3 Ultra (80-core GPU) variants: 512 GB
This chip is part of a family. Compare all M3 Ultra (80-core GPU) RAM variants →
Raw benchmark rows for M3 Ultra (80-core GPU, 256 GB)
Rows sorted by avg tok/s descending. Click source badge to see original measurement page.
| Model | Quant | Avg tok/s | Runtime | Source |
|---|---|---|---|---|
| Llama 3.2 1B | Q4_K - Medium | 177.9 tok/s | llamafile | ref |
| Llama 3.1 8B | Q4_K - Medium | 63.3 tok/s | llamafile | ref |
| Qwen 2.5 14B | Q4_K - Medium | 36.7 tok/s | llamafile | ref |
Models tested on this chip
Next step
Use Rankings when you need the best overall answer for a Mac, and keep this page open when you need the evidence behind that recommendation.
If you are comparing local Apple Silicon against rented cloud hardware, use AI Data Center Index for current GPU rental context.
Data
benchmarks.json — full dataset · chips.json — chip summaries · benchmarks.csv — CSV export
Data from in-house lab measurements plus community-published benchmarks. See all chip families →