Run on your Mac

Run Outlier Nano on Mac mini M4

Last updated 2026-06-18 · Outlier v1.11.469

What does Outlier Nano run like on Mac mini M4?

The pairing under examination is Outlier Nano (2.37 GB MLX 4-bit, Qwen3.5-4B base, 32K default context, 6 GB unified memory minimum) on a Mac mini M4 (Apple M4, 10 CPU cores, 10 GPU cores, 16|24|32 GB unified memory, 120 GB/s, 2024). Outlier loads this tier as one mlx_lm process inside the bundled FastAPI sidecar. Apple Silicon decode on dense 4-bit weights is bandwidth-bound, so the ceiling on this exact Mac mini M4 scales as the ratio 120 / 800 = 0.15× the published M1 Ultra number.

How much memory does the model take during generation?

Outlier’s Outlier Nano has a measured peak generation footprint of about 3 GB. The Mac mini M4 base configuration ships with 16|24|32 GB unified, and the rule of thumb is to leave roughly 4 GB for the OS and one open browser tab. On Mac mini M4 the Outlier Nano bandwidth-scaled estimate is ~10.8 tok/s [estimated from family-bandwidth ratio]; derivation: 71.7 tok/s on M1 Ultra × (120 / 800) bus ratio for the Apple M4. Treat as a first-order projection — the Outlier Nano number on this 2024 machine has not been formally measured.

What is the install path on Mac mini M4 for Outlier Nano?

The 2024 Mac mini M4 ships with 16|24|32 GB of unified memory, so headroom for the 6 GB Outlier Nano requirement is 10 GB on the base SKU. Step-by-step install instructions live on the install guide; the part that varies for this Apple M4 machine is that the 2.37 GB pull from Outlier-Ai/Outlier-Nano-4B-MLX-4bit over HTTPS lands in ~/Library/Application Support/Outlier/models/nano/ and download time is bandwidth-limited by the network, not by the 10-core Apple M4 CPU.

What context window can I use on a 16 GB Mac?

The Outlier Nano tier defaults to 32K context and caps at 256K. KV cache scales linearly with context length on dense models, so longer contexts trade headroom for capacity. On a 16|24|32 GB Mac mini M4, the default context is the safe starting point.

What works well on this pairing, and what is still rough?

• The Mac mini M4 (Apple M4, 120 GB/s) is bandwidth-bound on dense 4-bit decode; the throughput ceiling for Outlier Nano on this exact machine is approximately 120 divided by 2.37 GB tokens per second.
• The 16 GB base SKU of the Mac mini M4 swaps aggressively under the Outlier Nano tier; close Chrome and other RAM-heavy applications before generation.
• The Apple M4 (120 GB/s, 10 GPU cores) is current Apple Store silicon as of 2025; new-purchase tok-per-second-per-dollar is best in the lineup at the Outlier Nano tier.
• Cold boot of Outlier on a 10-core Apple M4 is roughly 50 seconds while the bundled Python.framework warms; this is documented on the v1.9 backlog under the boot-time investigation thread.

What is the unique number for Outlier Nano on Mac mini M4?

2.37 GB on disk against 16|24|32 GB unified memory means the weights alone consume about 15% of the base SKU’s RAM on the Mac mini M4.

Should I pick a different tier?

For heavier work, the Lite tier is the next step up; check that the 16|24|32 GB SKU you have can fit its weights before installing. The unified-memory explainer works the bandwidth math out, and the bandwidth ratio for Mac mini M4 is 0.15× M1 Ultra reference.

Download Outlier for Mac

Requires Apple Silicon (M1, M2, M3, or M4) — Intel Macs are not supported. macOS 12+.