Run Apertus 1.5 70B on M4 Ultra

Yes — Apertus 1.5 70B (70B) runs at 20 tok/s on M4 Ultra with 96 GB RAM using Q4_K_M quantization via MLX. First token latency is 0.6s. A capable open-source LLM with 70B parameters.

Speed
20
tok/s
First Token
0.6
seconds
RAM Needed
96
GB minimum
Engine
MLX
recommended
Want to run Apertus 1.5 70B faster — or step up to bigger models? Find your Mac →

Benchmark Details

The LLMCheck index estimates Apertus 1.5 70B on M4 Ultra using our published methodology: Q4_K_M quantization, memory-bandwidth scaling, and cross-referenced third-party benchmarks where available. Figures are transparent estimates — own this config? Submit a real benchmark →

MetricValue
Tokens per second20 tok/s
Time to first token0.6s
QuantizationQ4_K_M
Minimum RAM96 GB
Recommended engineMLX
Parameters70B
Benchmark date2026-08

Q4_K_M 70B MLX M4 Ultra

Setup Guide: Run Apertus 1.5 70B on M4 Ultra

The recommended engine for Apertus 1.5 70B on M4 Ultra is MLX. Install with pip and pull the model:

pip install mlx-lm
mlx_lm.generate --model mlx-community/apertus-15-70b-q4_k_m --prompt "Hello!"

Alternatively, you can use Ollama for a simpler setup:

ollama run apertus-15-70b

Performance on Other Apple Silicon Chips

ChipSpeedFirst TokenMin RAMEngine
M5 Max 11 tok/s 0.9s 128 GB MLX

System Requirements

To run Apertus 1.5 70B on M4 Ultra you need:

🛒 Get a Mac that runs Apertus 1.5 70B

Apertus 1.5 70B needs about 96 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:

As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep our benchmarks free.

Compare More Models

See how Apertus 1.5 70B stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard