Can I Run Poolside Laguna XS 2.1 on Mac Studio (M2 Ultra, 192 GB)?

Written by Jakub Rusinowski · Last updated August 15, 2026

Mac Studio (M2 Ultra, 192 GB) — what it gives a model

Usable memory for models	144 GB
Memory bandwidth	800 GB/s
Form factor	Desktop
Operating system	macOS
Memory upgradeable	No — soldered

Poolside Laguna XS 2.1 on Mac Studio (M2 Ultra, 192 GB): memory by quantization

Quant	Memory needed	Fits 144 GB?	Max context	Est. speed	Download
F16	68.6 GB	✓ Yes	256K	~53.2 tok/s	66 GB
Q8_0	37.7 GB	✓ Yes	256K	~80.8 tok/s	35.1 GB
Q6_K	29.7 GB	✓ Yes	256K	~93.3 tok/s	27.1 GB
Q5_K_M	26 GB	✓ Yes	256K	~100.5 tok/s	23.4 GB
Q4_K_M	22.6 GB	✓ Yes	256K	~108.3 tok/s	19.9 GB
Q3_K_M	16.7 GB	✓ Yes	256K	~124.7 tok/s	14.1 GB
Q2_K	13.5 GB	✓ Yes	256K	~136 tok/s	10.8 GB

What to watch out for

• This model does not publish its full attention configuration, so the KV-cache share of these figures is inferred from its parameter count rather than computed exactly.
• Memory on this machine is not upgradeable, so the configuration you buy is the ceiling for every model you will ever run on it.

Mac Studio M2 Ultra 192 GB limitations

• Superseded by the M3 Ultra, which is why it is often the better used buy for large-model work.
• Memory is soldered; the configuration chosen at purchase is permanent.

Recommended setup

Ollama or LM Studio (Metal) — MLX for the fastest Apple-native throughput

How these numbers are calculated

• Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
• 192 GB unified memory at 800 GB/s, shared between CPU and GPU.
• macOS reserves a share of unified memory for the system, so not all of it is available to a model.
• Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
• KV cache is inferred from the parameter count because this model does not publish its full attention configuration.

FAQ

Can I run Poolside Laguna XS 2.1 on Mac Studio (M2 Ultra, 192 GB)?

Yes, comfortably — Laguna XS 2.1 33B-A3B at Q8_0 needs about 37.7 GB of the 144 GB usable on Mac Studio (M2 Ultra, 192 GB), leaving ~106.3 GB spare and running at ~80.8 tok/s (estimated), with room for about 262,144 tokens of context.

Which quantization of Poolside Laguna XS 2.1 should I use on Mac Studio (M2 Ultra, 192 GB)?

Q8_0 — it needs about 37.7 GB of the 144 GB available, downloads as roughly 35.1 GB, and runs at an estimated 80.8 tokens/sec with up to 256K of context.

What limits Poolside Laguna XS 2.1 on Mac Studio (M2 Ultra, 192 GB)?

Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.

Which runtime should I use?

Ollama or LM Studio (Metal) — MLX for the fastest Apple-native throughput

Other Computers

• Poolside Laguna XS 2.1 on RTX 4090 desktop
• Poolside Laguna XS 2.1 on RTX 5090 desktop
• Poolside Laguna XS 2.1 on Mac Studio M3 Ultra 256 GB
• Poolside Laguna XS 2.1 on RTX 3090 desktop

Other Models on Mac Studio (M2 Ultra, 192 GB)

• Qwen 2.5 Family on Mac Studio (M2 Ultra, 192 GB)
• Qwen 2.5 VL on Mac Studio (M2 Ultra, 192 GB)
• Qwen 3 on Mac Studio (M2 Ultra, 192 GB)
• Qwen 3.5 on Mac Studio (M2 Ultra, 192 GB)
• Qwen 3.6 on Mac Studio (M2 Ultra, 192 GB)

Poolside Laguna XS 2.1 on GPUs

• Poolside Laguna XS 2.1 on NVIDIA GeForce RTX 5090
• Poolside Laguna XS 2.1 on NVIDIA GeForce RTX 5080
• Poolside Laguna XS 2.1 on NVIDIA GeForce RTX 5070 Ti
• Poolside Laguna XS 2.1 on NVIDIA GeForce RTX 5070

What This Model Is Good At

• Best local LLMs for coding
• Best local LLMs for rag
• Best local LLMs for agents

Model & Tools

• Poolside Laguna XS 2.1 model page
• VRAM calculator
• Check your own hardware

← Can I Run It? | Poolside Laguna XS 2.1 model page | Check your hardware