CanItRun Logocanitrun.

Apple M1 Ultra (64GB)

The Apple M1 Ultra (64GB) has 64 GB VRAM and 800 GB/s memory bandwidth. It can run 54 of our 70 tracked models natively in VRAM at 8k context.

With 64 GB LPDDR5, the Apple M1 Ultra (64GB) is a workstation-tier GPU that can run 54 models natively. It handles 70B-class models at Q4 quantization.

Apple M1 Ultra (64GB): 64GB at 800 GB/s — base M1 Ultra.

32B at Q4 native. 70B at Q2-Q3. ~5-8 t/s for 32B.

Same as 128GB.

VendorApple
ArchitectureApple M1 Ultra
CPU cores20 (16P + 4E)
VRAM64 GB (unified)
Memory typeLPDDR5
Memory bandwidth800 GB/s
Compute backendMETAL
TierWorkstation
Released2022
Models (native)54 / 70
Models (offload)0 / 70
Software: MLX gives the best performance on Apple Silicon; llama.cpp Metal backend is a solid alternative. Both are well-supported by Ollama.

Popular models for this GPU

Models this GPU runs natively in VRAM (54)

Too large for this GPU (16)

Frequently asked questions

How much VRAM does the Apple M1 Ultra (64GB) have?
The Apple M1 Ultra (64GB) has 64 GB of LPDDR5 with 800 GB/s memory bandwidth (unified system memory, shared between CPU and GPU).
What is the Apple M1 Ultra (64GB) best for?
With 64 GB of VRAM, the Apple M1 Ultra (64GB) is ideal for running 70B-class models at Q4 quantization and large MoE models — a workstation sweet spot for local inference.
What LLMs can the Apple M1 Ultra (64GB) run locally?
The Apple M1 Ultra (64GB) can run 54 of the 70 open-weight models tracked by CanItRun natively in VRAM at 8k context. Top options include: Llama 3.3 70B Instruct at Q5_K_M, Llama 3.1 8B Instruct at FP32, Llama 3.2 3B Instruct at FP32.
Can the Apple M1 Ultra (64GB) run Llama 3.3 70B Instruct?
Yes. The Apple M1 Ultra (64GB) runs Llama 3.3 70B Instruct natively in VRAM at Q5_K_M quantization, achieving approximately 17.7 tokens per second.
Can the Apple M1 Ultra (64GB) run Qwen 3.6 27B?
Yes. The Apple M1 Ultra (64GB) runs Qwen 3.6 27B natively in VRAM at Q8_0 quantization, achieving approximately 29.6 tokens per second.
Can the Apple M1 Ultra (64GB) run Llama 3.1 8B Instruct?
Yes. The Apple M1 Ultra (64GB) runs Llama 3.1 8B Instruct natively in VRAM at FP32 quantization, achieving approximately 25 tokens per second.