Models

Ornith 1.5 397B

ornith-ai · August 2026

Get

Hugging Face

A 403B, 18B active chat model from ornith-ai that reads pictures as well as text.

Downloads: 1.3M
Likes: 64
Parameters: 403B, 18B active
Context: 256K
Licence: mit

Too big for this Mac

Its smallest size needs 86 GB of memory, more than M4 · 16 GB can give it.

Ornith 1.5 397B runs well from a Mac with an M1 Ultra and 128 GB of memory. 10 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From bartowski/Ornith-1.5-397B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

Size	Download	Memory at 8K	Get
IQ1_S	82 GB	82 GBestimated	Open
IQ1_M	91 GB	92 GBestimated	Open
IQ2_XXS	106 GB	107 GBestimated	Open
IQ2_XS	118 GB	119 GBestimated	Open
IQ2_S	121 GB	121 GBestimated	Open
IQ2_M	133 GB	133 GBestimated	Open
Q2_K	139 GB	140 GBestimated	Open
Q2_K_L	140 GB	141 GBestimated	Open
IQ3_XXS	166 GB	166 GBestimated	Open
Q3_K_S	173 GB	173 GBestimated	Open
IQ3_XS	181 GB	182 GBestimated	Open
Q3_K_M	181 GB	182 GBestimated	Open
Q3_K_L	189 GB	190 GBestimated	Open
IQ3_M	190 GB	190 GBestimated	Open
Q3_K_XL	190 GB	191 GBestimated	Open
IQ4_XS	212 GB	212 GBestimated	Open
IQ4_NL	225 GB	225 GBestimated	Open
Q4_0	225 GB	226 GBestimated	Open
Q4_K_S	233 GB	233 GBestimated	Open
Q4_K_M	242 GB	242 GBestimated	Open
Q4_1	249 GB	249 GBestimated	Open
Q5_K_S	274 GB	274 GBestimated	Open
Q5_K_M	283 GB	284 GBestimated	Open
Q8_0	422 GB	422 GBestimated	Open

Run It Anywhere

Ornith 1.5 397B is an open model, so any app that runs GGUF files can run it. These take the IQ1_S size from bartowski straight from Hugging Face.

LM StudioMac, Windows, LinuxOpen
JanMac, Windows, LinuxOpen
MstyMac, Windows, LinuxOpen
OllamaMac, Windows, Linuxollama run hf.co/bartowski/Ornith-1.5-397B-GGUF:IQ1_S
llama.cppMac, Windows, Linuxllama-server -hf bartowski/Ornith-1.5-397B-GGUF:IQ1_S
Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/Ornith-1.5-397B-GGUF:IQ1_S

On a PC Graphics Card

Video memory	Largest size that fits	Memory at 8K
6 GB	None worth running	needs 166 GB
8 GB	None worth running	needs 166 GB
10 GB	None worth running	needs 166 GB
11 GB	None worth running	needs 166 GB
12 GB	None worth running	needs 166 GB
16 GB	None worth running	needs 166 GB
20 GB	None worth running	needs 166 GB
24 GB	None worth running	needs 166 GB
32 GB	None worth running	needs 166 GB
48 GB	None worth running	needs 166 GB

The largest size, of those that keep three bits a weight or more, whose memory at an 8K chat is under 92% of the card's; the rest is the desktop's. Estimated from the files, the same way as for a Mac; speed on a PC is not estimated yet. A size that doesn't fit the card still runs, slower, with part of it in ordinary memory.

On every Mac

Mac	Size	Fit	Memory	Writes
M6 · 16 GB	IQ1_S	Too big	86 GB
M6 · 24 GB	IQ1_S	Too big	86 GB
M6 · 32 GB	IQ1_S	Too big	86 GB
M5 · 16 GB	IQ1_S	Too big	86 GB
M5 · 24 GB	IQ1_S	Too big	86 GB
M5 · 32 GB	IQ1_S	Too big	86 GB
M5 Pro · 24 GB	IQ1_S	Too big	86 GB
M5 Pro · 48 GB	IQ1_S	Too big	86 GB
M5 Pro · 64 GB	IQ1_S	Too big	86 GB
M5 Max · 36 GB	IQ1_S	Too big	86 GB
M5 Max · 48 GB	IQ1_S	Too big	86 GB
M5 Max · 64 GB	IQ1_S	Too big	86 GB
M5 Max · 128 GB	IQ1_M	Runs well	96 GB	39
M5 Ultra · 96 GB	IQ1_S	Too big	86 GB
M5 Ultra · 256 GB	Q3_K_S	Runs great	174 GB	44
M5 Ultra · 512 GB	Q6_K	Runs great	336 GB	28
M4 · 16 GB	IQ1_S	Too big	86 GB
M4 · 24 GB	IQ1_S	Too big	86 GB
M4 · 32 GB	IQ1_S	Too big	86 GB
M4 Pro · 24 GB	IQ1_S	Too big	86 GB
M4 Pro · 48 GB	IQ1_S	Too big	86 GB
M4 Pro · 64 GB	IQ1_S	Too big	86 GB
M4 Max · 36 GB	IQ1_S	Too big	86 GB
M4 Max · 48 GB	IQ1_S	Too big	86 GB
M4 Max · 64 GB	IQ1_S	Too big	86 GB
M4 Max · 96 GB	IQ1_S	Too big	86 GB
M4 Max · 128 GB	IQ1_M	Runs well	96 GB	24
M3 · 8 GB	IQ1_S	Too big	86 GB
M3 · 16 GB	IQ1_S	Too big	86 GB
M3 · 24 GB	IQ1_S	Too big	86 GB
M3 Pro · 18 GB	IQ1_S	Too big	86 GB
M3 Pro · 36 GB	IQ1_S	Too big	86 GB
M3 Max · 36 GB	IQ1_S	Too big	86 GB
M3 Max · 48 GB	IQ1_S	Too big	86 GB
M3 Max · 64 GB	IQ1_S	Too big	86 GB
M3 Max · 96 GB	IQ1_S	Too big	86 GB
M3 Max · 128 GB	IQ1_M	Runs well	96 GB	21
M3 Ultra · 96 GB	IQ1_S	Too big	86 GB
M3 Ultra · 256 GB	Q3_K_S	Runs great	174 GB	23
M3 Ultra · 512 GB	Q6_K	Runs great	336 GB	16
M2 · 8 GB	IQ1_S	Too big	86 GB
M2 · 16 GB	IQ1_S	Too big	86 GB
M2 · 24 GB	IQ1_S	Too big	86 GB
M2 Pro · 16 GB	IQ1_S	Too big	86 GB
M2 Pro · 32 GB	IQ1_S	Too big	86 GB
M2 Max · 32 GB	IQ1_S	Too big	86 GB
M2 Max · 64 GB	IQ1_S	Too big	86 GB
M2 Max · 96 GB	IQ1_S	Too big	86 GB
M2 Ultra · 64 GB	IQ1_S	Too big	86 GB
M2 Ultra · 128 GB	IQ1_M	Runs well	96 GB	31
M2 Ultra · 192 GB	IQ2_S	Runs great	125 GB	27
M1 · 8 GB	IQ1_S	Too big	86 GB
M1 · 16 GB	IQ1_S	Too big	86 GB
M1 Pro · 16 GB	IQ1_S	Too big	86 GB
M1 Pro · 32 GB	IQ1_S	Too big	86 GB
M1 Max · 32 GB	IQ1_S	Too big	86 GB
M1 Max · 64 GB	IQ1_S	Too big	86 GB
M1 Ultra · 64 GB	IQ1_S	Too big	86 GB
M1 Ultra · 128 GB	IQ1_M	Runs well	96 GB	23

Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.

Packaged by

2 repositories on Hugging Face carry this model for llama.cpp. Outmake reads them all as one model.

Information

Maker: ornith-ai
Released: August 2026
Parameters: 403B, 18B active
Longest chat: 256K tokens
Reads pictures: Yes
Licence: mit
Downloads this month: 1,261,164
Likes: 64
Sizes: 24, 82 to 422 GB
Kind: Maker's release
Source: ornith-ai/Ornith-1.5-397B-GGUF

Badge for the model card

Shows on Ornith 1.5 397B's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Files, downloads and likes read from Hugging Face on November 5, 2026 and refreshed every morning. Memory and speed are estimates until someone running Outmake measures this model on that Mac.