Apple M1 Ultra for local LLMs
Written by Jakub Rusinowski · Last updated
With 128 GB at 800 GB/s, the M1 Ultra runs 140 catalogued models at Q4_K_M with 8K context. The largest that fits is Mistral Medium 3.5 128B (~80.9 GB), and the top pick is Qwen 3.5 122B-A10B at 28–58 tok/s.
Two M1 Max dies fused together. Up to 128 GB unified memory at 800 GB/s — comfortably runs 70B models at Q4. Mac Studio exclusive.
Models that run on the M1 Ultra
Q4_K_M, 8K context, 96 GB usable. Ranked by quality and speed.
| Model | Memory · marker = 96 GB | VRAM | Speed |
|---|---|---|---|
| Qwen 3.5 122B-A10B Qwen 3.5 | 74.5 GB | 28–58 tok/s | |
| Nemotron 3 Super 120B-A12B Nemotron 3 Super | 73.3 GB | 25–52 tok/s | |
| Nemotron 3.5 Super VL 120B-A12B Nemotron 3.5 | 73.3 GB | 25–52 tok/s | |
| Mistral Small 4 119B-A6.5B Mistral Small 4 | 72.6 GB | 16–34 tok/s | |
| GPT-OSS 120B GPT-OSS | 71.6 GB | 60–125 tok/s | |
| Llama 4.5 Scout Llama 4.5 | 66.6 GB | 20–41 tok/s | |
| Llama 4 Scout 17B Llama 4 | 68.2 GB | 22–45 tok/s | |
| Qwen3-Coder-Next (80B-A3B MoE) Qwen3-Coder | 49.4 GB | 84–174 tok/s | |
| Kolibri 1 Kolibri | 48.5 GB | 78–161 tok/s | |
| Nemotron 3.5 Bolt 74B Nemotron 3.5 | 45.6 GB | 42–87 tok/s | |
| Cosmos 3 Super Cosmos 3 | 39.4 GB | 12–26 tok/s | |
| K2 Horizon MoVA 36B-A4B K2 Horizon | 25 GB | 53–111 tok/s |
Buy it or rent the same memory
Buy the card, or rent a GPU with the same memory by the hour to try models first.
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
Speed vs other GPUs
Llama 3.1 8B, Q4_K_M. Estimated ranges. How this is calculated
Specifications
Specs last updated 2026-07-12.
- Memory
- 128 GB
- Memory bandwidth
- 800 GB/s
- Architecture
- ARM, 5nm TSMC
- Series
- Apple Silicon
- Board power
- 60 W
- Release year
- 2022
- Usable for models
- 96 GB (~75% of unified)
Similar GPUs
Frequently asked questions
Can the Apple M1 Ultra run local LLMs?
Yes. With 128 GB (96 GB usable by a model) it runs 140 of the catalogued models at Q4_K_M with 8K context; the largest is Mistral Medium 3.5 128B, needing about 80.9 GB.
How fast is the Apple M1 Ultra for AI inference?
It is estimated to run Llama 3.1 8B at 43–89 tok/s at Q4_K_M. Llama 3.3 70B is estimated at 6.3–13 tok/s. These are modelled estimates from memory bandwidth, not measurements; the methodology page shows the formula.
What LLMs can I run on 128 GB?
Among the best that fit: Qwen 3.5 122B-A10B, Nemotron 3 Super 120B-A12B, Nemotron 3.5 Super VL 120B-A12B, Mistral Small 4 119B-A6.5B, GPT-OSS 120B. The quickest start is Ollama: ollama run llama3.3:70b.
Keep going
Can I run it on the M1 Ultra?
More for this card
Tools