Mac Studio M3 Ultra 512GB
512GB Unified memory
/
819 GB/s
/
100W
/
$7,499 MSRP
/
Released 2025
Overview
Apple’s 2025 workstation built on the M3 Ultra (March 2025 announcement), the first Mac with 512GB of unified memory. Three configurations: 96GB at $3,999, 256GB at $5,499, 512GB at $7,499. All share 819 GB/s of memory bandwidth and a 100W-class chip.
What it does well:
- 512GB in one box: at launch it was the only consumer machine that could hold DeepSeek R1 or Kimi-class checkpoints whole, no offloading.
- Efficiency: a full inference node at 100W of chip power - roughly a tenth of a comparable multi-GPU server’s draw.
Where it falls short: MLX and GGUF runtimes, no CUDA; 819 GB/s is well below the M4/M5 Ultra tiers, so tok/s on large models trails them. Apple has superseded it twice since, and the M5 Ultra beats it on every spec.
Run it locally: LM Studio, Ollama, or mlx-lm. Check the modeldex results for fit and estimated tok/s per checkpoint on your memory tier.
Chip specs
- Apple M3 Ultra
- 512GB unified
- 819 GB/s
- 2025
- Moderate
Top models (114 compatible)
GLM-5.3-Flash
UD-IQ1_S
Runs comfortably
93.1GB / 100.0GB min
94
ⓘ estimated
DeepSeek V4.1 Flash
NVFP4
Runs comfortably
260.0GB / 265.0GB min
65
ⓘ estimated
DeepSeek V4 Flash 0731
UD-Q4_K_XL
Runs comfortably
155.0GB / 168.0GB min
69
ⓘ estimated
Ornith-1.5-397B
Q4_K_M
Runs comfortably
244.0GB / 248.0GB min
2
ⓘ estimated
Qwen3.8-Flash-Next
UD-IQ1_S
Runs comfortably
72.5GB / 78.0GB min
141
ⓘ estimated