Outmake Models
Models

Qwythos 9B V2

empero-ai · July 2026

Get Hugging Face

A 9.2B chat model from empero-ai that reads pictures as well as text.

Downloads
308K
Likes
325
Parameters
9.2B
Context
1M
Licence
apache-2.0

Runs great on M4 · 16 GB

The Q6_K size from empero-ai is the one to get: a 7.5 GB download that uses 7.7 GB of memory, and writes about 7.8 words a second.

Qwythos 9B V2 runs well from a Mac with an M1 and 8 GB of memory. 59 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From empero-ai/Qwythos-9B-v2-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q4_K_M5.7 GB6.0 GBestimatedOpen
Q5_K_M6.5 GB6.8 GBestimatedOpen
Q6_K7.5 GB7.7 GBestimatedOpen
Q8_09.5 GB9.8 GBestimatedOpen
BF1618 GB18 GBestimatedOpen

Run It Anywhere

Qwythos 9B V2 is an open model, so any app that runs GGUF files can run it. These take the Q6_K size from empero-ai straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/empero-ai/Qwythos-9B-v2-GGUF:Q6_K
  • llama.cppMac, Windows, Linuxllama-server -hf empero-ai/Qwythos-9B-v2-GGUF:Q6_K
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/empero-ai/Qwythos-9B-v2-GGUF:Q6_K
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 6.0 GB
8 GBQ6_K7.5 GB7.7 GB
10 GBQ8_09.5 GB9.8 GB
11 GBQ8_09.5 GB9.8 GB
12 GBQ8_09.5 GB9.8 GB
16 GBQ8_09.5 GB9.8 GB
20 GBBF1618 GB18 GB
24 GBBF1618 GB18 GB
32 GBBF1618 GB18 GB
48 GBBF1618 GB18 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
MacSizeFitMemoryWrites
M6 · 16 GBQ6_KRuns great7.7 GB12
M6 · 24 GBQ8_0Runs great11 GB11
M6 · 32 GBQ8_0Runs great15 GB11
M5 · 16 GBQ6_KRuns great7.7 GB10
M5 · 24 GBQ8_0Runs great11 GB9.6
M5 · 32 GBQ8_0Runs great15 GB9.6
M5 Pro · 24 GBQ8_0Runs great11 GB19
M5 Pro · 48 GBBF16Runs great21 GB11
M5 Pro · 64 GBBF16Runs great28 GB11
M5 Max · 36 GBQ8_0Runs great15 GB28
M5 Max · 48 GBBF16Runs great21 GB22
M5 Max · 64 GBBF16Runs great28 GB22
M5 Max · 128 GBBF16Runs great39 GB22
M5 Ultra · 96 GBBF16Runs great39 GB42
M5 Ultra · 256 GBBF16Runs great39 GB42
M5 Ultra · 512 GBBF16Runs great39 GB42
M4 · 16 GBQ6_KRuns great7.7 GB7.8
M4 · 24 GBQ8_0Runs great11 GB7.3
M4 · 32 GBQ8_0Runs great15 GB7.3
M4 Pro · 24 GBQ8_0Runs great11 GB16
M4 Pro · 48 GBBF16Runs great21 GB9.2
M4 Pro · 64 GBBF16Runs great28 GB9.2
M4 Max · 36 GBQ8_0Runs great15 GB22
M4 Max · 48 GBBF16Runs great21 GB17
M4 Max · 64 GBBF16Runs great28 GB17
M4 Max · 96 GBBF16Runs great39 GB13
M4 Max · 128 GBBF16Runs great39 GB17
M3 · 8 GBQ3_K_LRuns well5.3 GB9.1
M3 · 16 GBQ6_KRuns great7.7 GB6.6
M3 · 24 GBQ8_0Runs great11 GB6.1
M3 Pro · 18 GBQ6_KRuns great8.2 GB9.8
M3 Pro · 36 GBQ8_0Runs great15 GB9
M3 Max · 36 GBQ8_0Runs great15 GB17
M3 Max · 48 GBBF16Runs great21 GB13
M3 Max · 64 GBBF16Runs great28 GB13
M3 Max · 96 GBBF16Runs great39 GB10
M3 Max · 128 GBBF16Runs great39 GB13
M3 Ultra · 96 GBBF16Runs great39 GB25
M3 Ultra · 256 GBBF16Runs great39 GB25
M3 Ultra · 512 GBBF16Runs great39 GB25
M2 · 8 GBQ3_K_LRuns well5.3 GB9
M2 · 16 GBQ6_KRuns great7.7 GB6.6
M2 · 24 GBQ8_0Runs great11 GB6.1
M2 Pro · 16 GBQ6_KRuns great7.7 GB13
M2 Pro · 32 GBQ8_0Runs great15 GB12
M2 Max · 32 GBQ8_0Runs great15 GB22
M2 Max · 64 GBBF16Runs great28 GB13
M2 Max · 96 GBBF16Runs great39 GB13
M2 Ultra · 64 GBBF16Runs great28 GB25
M2 Ultra · 128 GBBF16Runs great39 GB25
M2 Ultra · 192 GBBF16Runs great39 GB25
M1 · 8 GBQ3_K_LRuns well5.3 GB6.1
M1 · 16 GBQ6_KRuns great7.7 GB4.5
M1 Pro · 16 GBQ6_KRuns great7.7 GB12
M1 Pro · 32 GBQ8_0Runs great15 GB11
M1 Max · 32 GBQ8_0Runs great15 GB20
M1 Max · 64 GBBF16Runs great28 GB13
M1 Ultra · 64 GBBF16Runs great28 GB23
M1 Ultra · 128 GBBF16Runs great39 GB23

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
empero-ai
Released
July 2026
Parameters
9.2B
Longest chat
1M tokens
Reads pictures
Yes
Licence
apache-2.0
Downloads this month
308,166
Likes
325
Sizes
5, 5.7 to 18 GB
Kind
Maker's release

Badge for the model card

Shows on Qwythos 9B V2's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Runs on Mac badge

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.

Your Computer

Every model is judged for this Mac. Both are under About This Mac, in the Apple menu.

Every Mac · Every graphics card

Civitai

Open in OutmakeHugging Face

New to Outmake? Download the free beta