Ornith 1.0 9B

Ornith 1.0 9B loads in 5.6 GB at Q4_K_M and is comfortable from 12 GB of memory, Macs included from the Mac Mini M6 16GB. Below: the full memory math, the cheapest card that runs it well, and the fastest machine we track.

PARAMETERS
9B
FORMAT
Q4_K_M
MIN MEMORY
8 GB
BEST FOR
Agentic coding on small machines

Memory math

Weights (Q4_K_M)
5.6 GB
+ KV cache (16k)
~2.0 GB
Total at 16k
~7.6 GB
Comfortable from
12 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

Cheapest GPU
~67 tok/s est. — ~$550 used
Fastest
~131 tok/s est.
Runs on a Mac?
~25 tok/s est.

More Ornith models

Ornith 1.0 35B

Frequently asked questions

How much memory does Ornith 1.0 9B need?

5.6 GB for the Q4_K_M weights, plus ~2.0 GB of KV-cache at 16k context — about 7.6 GB total. Comfortable from 12 GB of VRAM or unified memory.

Does Ornith 1.0 9B run on a Mac?

Yes — from the Mac Mini M6 16GB (~25 tok/s est.). Unified memory means the RAM budget is the only limit.

What is the cheapest GPU for Ornith 1.0 9B?

The AMD Radeon RX 7900 XT is the cheapest tracked card that runs Ornith 1.0 9B comfortably — ~67 tok/s est. at ~$550 used.

Cite this page

ModelFit: Ornith 1.0 9B — specs, memory math and hardware verdicts.
https://modelfit.io/models/ornith-9b/ (dataset updated 2026-09-03, CC BY 4.0).