Models

Ling 3.1 Flash

Ant Group · October 2026

Get

Hugging Face

A 572B, 25B active chat model from Ant Group.

Downloads: 8.4K
Likes: 12
Parameters: 572B, 25B active
Context: 256K
Licence: mit

Too big for this Mac

Its smallest size needs 133 GB of memory, more than M4 · 16 GB can give it.

Ratings

Expert questionsHumanity's Last Exam: 37.4%

Humanity's Last Exam: questions written to be beyond today's models. Scores are the ones filed on the model's Hugging Face page or the board's own table, read 2026-11-05; some come from the maker.

Ling 3.1 Flash runs well from a Mac with an M2 Ultra and 192 GB of memory. 5 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From bartowski/Ling-3.1-flash-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

Size	Download	Memory at 8K	Get
IQ1_S	131 GB	131 GBestimated	Open
IQ1_M	145 GB	145 GBestimated	Open
IQ2_XXS	171 GB	171 GBestimated	Open
IQ2_XS	177 GB	178 GBestimated	Open
IQ2_S	188 GB	189 GBestimated	Open
IQ2_M	213 GB	213 GBestimated	Open
Q2_K	219 GB	219 GBestimated	Open
IQ3_XXS	247 GB	248 GBestimated	Open
IQ3_XS	260 GB	261 GBestimated	Open
Q3_K_S	262 GB	262 GBestimated	Open
Q3_K_M	275 GB	275 GBestimated	Open
Q3_K_L	288 GB	288 GBestimated	Open
IQ3_M	297 GB	297 GBestimated	Open
IQ4_XS	317 GB	317 GBestimated	Open
Q4_0	327 GB	327 GBestimated	Open
Q4_K_S	337 GB	337 GBestimated	Open
Q4_K_M	358 GB	358 GBestimated	Open
IQ4_NL	358 GB	358 GBestimated	Open
Q4_1	360 GB	360 GBestimated	Open
Q5_K_S	406 GB	406 GBestimated	Open
Q5_K_M	436 GB	436 GBestimated	Open
Q6_K_S	479 GB	479 GBestimated	Open
Q6_K	502 GB	502 GBestimated	Open
Q8_0	608 GB	609 GBestimated	Open

Run It Anywhere

Ling 3.1 Flash 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/Ling-3.1-flash-GGUF:IQ1_S
llama.cppMac, Windows, Linuxllama-server -hf bartowski/Ling-3.1-flash-GGUF:IQ1_S
Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/Ling-3.1-flash-GGUF:IQ1_S

On a PC Graphics Card

Video memory	Largest size that fits	Memory at 8K
6 GB	None worth running	needs 248 GB
8 GB	None worth running	needs 248 GB
10 GB	None worth running	needs 248 GB
11 GB	None worth running	needs 248 GB
12 GB	None worth running	needs 248 GB
16 GB	None worth running	needs 248 GB
20 GB	None worth running	needs 248 GB
24 GB	None worth running	needs 248 GB
32 GB	None worth running	needs 248 GB
48 GB	None worth running	needs 248 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	133 GB
M6 · 24 GB	IQ1_S	Too big	133 GB
M6 · 32 GB	IQ1_S	Too big	133 GB
M5 · 16 GB	IQ1_S	Too big	133 GB
M5 · 24 GB	IQ1_S	Too big	133 GB
M5 · 32 GB	IQ1_S	Too big	133 GB
M5 Pro · 24 GB	IQ1_S	Too big	133 GB
M5 Pro · 48 GB	IQ1_S	Too big	133 GB
M5 Pro · 64 GB	IQ1_S	Too big	133 GB
M5 Max · 36 GB	IQ1_S	Too big	133 GB
M5 Max · 48 GB	IQ1_S	Too big	133 GB
M5 Max · 64 GB	IQ1_S	Too big	133 GB
M5 Max · 128 GB	IQ1_S	Too big	133 GB
M5 Ultra · 96 GB	IQ1_S	Too big	133 GB
M5 Ultra · 256 GB	IQ2_XXS	Runs great	173 GB	45
M5 Ultra · 512 GB	Q4_K_S	Runs great	338 GB	27
M4 · 16 GB	IQ1_S	Too big	133 GB
M4 · 24 GB	IQ1_S	Too big	133 GB
M4 · 32 GB	IQ1_S	Too big	133 GB
M4 Pro · 24 GB	IQ1_S	Too big	133 GB
M4 Pro · 48 GB	IQ1_S	Too big	133 GB
M4 Pro · 64 GB	IQ1_S	Too big	133 GB
M4 Max · 36 GB	IQ1_S	Too big	133 GB
M4 Max · 48 GB	IQ1_S	Too big	133 GB
M4 Max · 64 GB	IQ1_S	Too big	133 GB
M4 Max · 96 GB	IQ1_S	Too big	133 GB
M4 Max · 128 GB	IQ1_S	Too big	133 GB
M3 · 8 GB	IQ1_S	Too big	133 GB
M3 · 16 GB	IQ1_S	Too big	133 GB
M3 · 24 GB	IQ1_S	Too big	133 GB
M3 Pro · 18 GB	IQ1_S	Too big	133 GB
M3 Pro · 36 GB	IQ1_S	Too big	133 GB
M3 Max · 36 GB	IQ1_S	Too big	133 GB
M3 Max · 48 GB	IQ1_S	Too big	133 GB
M3 Max · 64 GB	IQ1_S	Too big	133 GB
M3 Max · 96 GB	IQ1_S	Too big	133 GB
M3 Max · 128 GB	IQ1_S	Too big	133 GB
M3 Ultra · 96 GB	IQ1_S	Too big	133 GB
M3 Ultra · 256 GB	IQ2_XXS	Runs great	173 GB	22
M3 Ultra · 512 GB	Q4_K_S	Runs great	338 GB	15
M2 · 8 GB	IQ1_S	Too big	133 GB
M2 · 16 GB	IQ1_S	Too big	133 GB
M2 · 24 GB	IQ1_S	Too big	133 GB
M2 Pro · 16 GB	IQ1_S	Too big	133 GB
M2 Pro · 32 GB	IQ1_S	Too big	133 GB
M2 Max · 32 GB	IQ1_S	Too big	133 GB
M2 Max · 64 GB	IQ1_S	Too big	133 GB
M2 Max · 96 GB	IQ1_S	Too big	133 GB
M2 Ultra · 64 GB	IQ1_S	Too big	133 GB
M2 Ultra · 128 GB	IQ1_S	Too big	133 GB
M2 Ultra · 192 GB	IQ1_S	Runs great	131 GB	25
M1 · 8 GB	IQ1_S	Too big	133 GB
M1 · 16 GB	IQ1_S	Too big	133 GB
M1 Pro · 16 GB	IQ1_S	Too big	133 GB
M1 Pro · 32 GB	IQ1_S	Too big	133 GB
M1 Max · 32 GB	IQ1_S	Too big	133 GB
M1 Max · 64 GB	IQ1_S	Too big	133 GB
M1 Ultra · 64 GB	IQ1_S	Too big	133 GB
M1 Ultra · 128 GB	IQ1_S	Too big	133 GB

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: Ant Group
Released: October 2026
Parameters: 572B, 25B active
Longest chat: 256K tokens
Reads pictures: No
Licence: mit
Downloads this month: 8,413
Likes: 12
Sizes: 24, 131 to 608 GB
Kind: Maker's release
Source: inclusionAI/Ling-3.1-flash

Badge for the model card

Shows on Ling 3.1 Flash'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.