Apple M4 Pro: 64 GB, what it runs and what it costs
Specs checked · Aliteq Team · How we compute this
The short answer
The Apple M4 Pro has 64 GB of VRAM and 273 GB/s of memory bandwidth (2024). At Q4_K_M quality or better with an 8K context it runs 37 of the open models we track; the largest that fits with room to spare is DeepSeek R1 Distill 70B. We don't see it for rent on the providers we track right now.
- VRAM
- 64 GB
- Bandwidth
- 273 GB/s
- Board power
- —
- Launched
- 2024
Source: Apple specification · bandwidth not yet confirmed by a second source, so we don't estimate speed from it
Buy it or rent it
Not for rent on the providers we track right now. See every card that is.
Which AI models run on the Apple M4 Pro
Computed by our VRAM engine for an 8K context: the best quality each model fits at in 64 GB. Speeds are upper-bound estimates from memory bandwidth, not benchmarks.
| Model | Best fit | VRAM used | ~tok/s |
|---|---|---|---|
| Llama 4 Scout 17B-16E (MoE) | Q4_K_M | 62 GB98% · tight | — |
| GLM-4.5-Air (MoE) | Q4_K_M | 62 GB97% · tight | — |
| Qwen3-Next 80B-A3B (MoE) | Q6_K | 62 GB97% · tight | — |
| Qwen3-Coder-Next 80B-A3B (MoE) | Q6_K | 62 GB97% · tight | — |
| DeepSeek R1 Distill 70B | Q6_K | 57 GB89% | — |
| Mixtral 8x7B (MoE) | Q8_0 | 48 GB75% | — |
| Qwen3.6 35B-A3B (MoE) | Q8_0 | 37 GB57% | — |
| Ornith-1.5 35B-A3B (MoE) | Q8_0 | 37 GB57% | — |
| Qwen3.5 35B-A3B (MoE) | Q8_0 | 37 GB57% | — |
| DeepSeek R1 Distill 32B | F16 | 64 GB100% · tight | — |
| Qwen2.5 Coder 32B | F16 | 64 GB100% · tight | — |
| Qwen2.5 32B Instruct | F16 | 64 GB100% · tight | — |
Fits only below Q4 quality: Qwen3.5 122B-A10B (MoE) (IQ4_XS), NVIDIA Nemotron 3 Super 120B-A12B (MoE) (IQ4_XS), gpt-oss 120B (MoE) (Q4_0).
Too big for 64 GB (rent a bigger card instead): what it costs to run Qwen3.8-Flash-Next (MoE), what it costs to run MiniMax-M2.5 (MoE), what it costs to run Qwen3 235B-A22B (MoE), what it costs to run MiniMax-M3 (MoE).
Check a model and context of your own in the VRAM calculator.
Cards to compare it with
- RTX A6000 · 48 GB
- AMD Radeon PRO W7900 · 48 GB
- RTX 6000 Ada Generation · 48 GB
- Apple M4 Max (14-core CPU / 32-core GPU) · 36 GB
- Apple M2 Max · 96 GB
- RTX 5000 Ada Generation · 32 GB
Every card by VRAM tier: all GPUs for gaming and local AI.