Ornith 1.0 35B

Ornith 1.0 35B loads in 21.2 GB at Q4_K_M, comfortably from 36 GB of memory — Macs included, starting at the MacBook Pro M5 Pro 48GB. Below is the memory math we ran for it, the cheapest card that fits, and the fastest machine we track.

PARAMETERS
35B
FORMAT
Q4_K_M
MIN MEMORY
32 GB
BEST FOR
Agentic coding

How much memory does Ornith 1.0 35B need?

Ornith 1.0 35B is a 35B dense model built for agentic coding. The weights alone take 21.2 GB at Q4_K_M; add context and the OS, and it belongs in the 36 GB memory class. Every number below comes from the ModelFit engine — speeds are estimates, labeled est., because we do not benchmark hardware ourselves.

Weights (Q4_K_M)
21.2 GB
+ KV cache (16k)
~4.0 GB
Total at 16k
~25.2 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).

Which machines run Ornith 1.0 35B?

Cheapest GPU
~25 tok/s est. — ~$700 used
Fastest
~41 tok/s est.
Runs on a Mac?
~11 tok/s est.

Verdicts come from the same engine that powers our fit checker — memory capacity first, bandwidth for speed. How we compute these numbers · Hugging Face model card

More Ornith models

Frequently asked questions

How much memory does Ornith 1.0 35B need?

21.2 GB for the Q4_K_M weights, plus ~4.0 GB of KV-cache at 16k context — about 25.2 GB total. Comfortable from 36 GB of VRAM or unified memory.

Does Ornith 1.0 35B run on a Mac?

Yes — from the MacBook Pro M5 Pro 48GB (~11 tok/s est.). Unified memory means the RAM budget is the only limit.

What is the cheapest GPU for Ornith 1.0 35B?

The AMD Radeon RX 7900 XTX is the cheapest tracked card that runs Ornith 1.0 35B comfortably — ~25 tok/s est. at ~$700 used.

Cite this page

ModelFit: Ornith 1.0 35B — specs, memory math and hardware verdicts.
https://modelfit.io/models/ornith-35b/ (dataset updated 2026-09-03, CC BY 4.0).