Laguna XS 2.1
Laguna XS 2.1 loads in 20.3 GB at Q4_K_M and is comfortable from 36 GB of memory, Macs included from the MacBook Pro M5 Pro 48GB. Below: the full memory math, the cheapest card that runs it well, and the fastest machine we track.
PARAMETERS
33B (3B active)
FORMAT
Q4_K_M
MIN MEMORY
32 GB
BEST FOR
Agentic coding, Long-horizon tasks
Memory math
Weights (Q4_K_M)
20.3 GB
+ KV cache (16k)
~4.0 GB
Total at 16k
~24.3 GB
Comfortable from
36 GB
KV = fp16 estimate (q8_0 cache roughly halves it). "Comfortable" = weights + KV within the engine's tiered budget (~70-85% of memory).
Hardware snapshot
More Laguna models
Frequently asked questions
How much memory does Laguna XS 2.1 need?
20.3 GB for the Q4_K_M weights, plus ~4.0 GB of KV-cache at 16k context — about 24.3 GB total. Comfortable from 36 GB of VRAM or unified memory.
Does Laguna XS 2.1 run on a Mac?
Yes — from the MacBook Pro M5 Pro 48GB (~40 tok/s est.). Unified memory means the RAM budget is the only limit.
What is the cheapest GPU for Laguna XS 2.1?
The AMD Radeon RX 7900 XTX is the cheapest tracked card that runs Laguna XS 2.1 comfortably — ~74 tok/s est. at ~$700 used.
Cite this page
ModelFit: Laguna XS 2.1 — specs, memory math and hardware verdicts. https://modelfit.io/models/laguna-xs-2.1/ (dataset updated 2026-09-03, CC BY 4.0).