Models

LongCat Flash Lite

meituan-longcat · January 2026

Get

Hugging Face

A 69B, 4.1B active chat model from meituan-longcat.

Downloads: 3.6K
Likes: 10
Parameters: 69B, 4.1B active
Context: 320K
Licence: mit

Too big for this Mac

Its smallest size needs 38 GB of memory, more than M4 · 16 GB can give it.

Ratings

KnowledgeMMLU-Pro: 78.3%

MMLU-Pro: graduate-level questions in 14 subjects. 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.

LongCat Flash Lite runs well from a Mac with an M1 Max and 64 GB of memory. 24 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From InquiringMinds-AI/LongCat-Flash-Lite-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

Size	Download	Memory at 8K	Get
Q3_K_L	33 GB	33 GBestimated	Open
Q4_K_M	40 GB	40 GBestimated	Open
Q5_K_M	48 GB	48 GBestimated	Open
Q6_K	56 GB	57 GBestimated	Open
Q8_0	73 GB	73 GBestimated	Open
BF16	137 GB	137 GBestimated	Open

Run It Anywhere

LongCat Flash Lite is an open model, so any app that runs GGUF files can run it. These take the Q3_K_L size from InquiringMinds-AI straight from Hugging Face.

LM StudioMac, Windows, LinuxOpen
JanMac, Windows, LinuxOpen
MstyMac, Windows, LinuxOpen
OllamaMac, Windows, Linuxollama run hf.co/InquiringMinds-AI/LongCat-Flash-Lite-GGUF:Q3_K_L
llama.cppMac, Windows, Linuxllama-server -hf InquiringMinds-AI/LongCat-Flash-Lite-GGUF:Q3_K_L
Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/InquiringMinds-AI/LongCat-Flash-Lite-GGUF:Q3_K_L

On a PC Graphics Card

Video memory	Largest size that fits	Download	Memory at 8K
6 GB	None worth running		needs 33 GB
8 GB	None worth running		needs 33 GB
10 GB	None worth running		needs 33 GB
11 GB	None worth running		needs 33 GB
12 GB	None worth running		needs 33 GB
16 GB	None worth running		needs 33 GB
20 GB	None worth running		needs 33 GB
24 GB	None worth running		needs 33 GB
32 GB	None worth running		needs 33 GB
48 GB	Q4_K_M	40 GB	40 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	Q3_K_L	Too big	38 GB
M6 · 24 GB	Q3_K_L	Too big	38 GB
M6 · 32 GB	Q3_K_L	Too big	38 GB
M5 · 16 GB	Q3_K_L	Too big	38 GB
M5 · 24 GB	Q3_K_L	Too big	38 GB
M5 · 32 GB	Q3_K_L	Too big	38 GB
M5 Pro · 24 GB	Q3_K_L	Too big	38 GB
M5 Pro · 48 GB	Q3_K_L	Tight fit	38 GB	38
M5 Pro · 64 GB	Q4_K_M	Runs well	42 GB	33
M5 Max · 36 GB	Q3_K_L	Too big	38 GB
M5 Max · 48 GB	Q3_K_L	Tight fit	38 GB	69
M5 Max · 64 GB	Q4_K_M	Runs well	42 GB	61
M5 Max · 128 GB	Q6_K	Runs great	61 GB	51
M5 Ultra · 96 GB	Q4_K_M	Runs great	45 GB	107
M5 Ultra · 256 GB	Q8_0	Runs great	78 GB	88
M5 Ultra · 512 GB	Q8_0	Runs great	78 GB	88
M4 · 16 GB	Q3_K_L	Too big	38 GB
M4 · 24 GB	Q3_K_L	Too big	38 GB
M4 · 32 GB	Q3_K_L	Too big	38 GB
M4 Pro · 24 GB	Q3_K_L	Too big	38 GB
M4 Pro · 48 GB	Q3_K_L	Tight fit	38 GB	28
M4 Pro · 64 GB	Q4_K_M	Runs well	42 GB	25
M4 Max · 36 GB	Q3_K_L	Too big	38 GB
M4 Max · 48 GB	Q3_K_L	Tight fit	38 GB	44
M4 Max · 64 GB	Q4_K_M	Runs well	42 GB	41
M4 Max · 96 GB	Q4_K_M	Runs great	45 GB	34
M4 Max · 128 GB	Q6_K	Runs great	61 GB	35
M3 · 8 GB	Q3_K_L	Too big	38 GB
M3 · 16 GB	Q3_K_L	Too big	38 GB
M3 · 24 GB	Q3_K_L	Too big	38 GB
M3 Pro · 18 GB	Q3_K_L	Too big	38 GB
M3 Pro · 36 GB	Q3_K_L	Too big	38 GB
M3 Max · 36 GB	Q3_K_L	Too big	38 GB
M3 Max · 48 GB	Q3_K_L	Tight fit	38 GB	38
M3 Max · 64 GB	Q4_K_M	Runs well	42 GB	35
M3 Max · 96 GB	Q4_K_M	Runs great	45 GB	28
M3 Max · 128 GB	Q6_K	Runs great	61 GB	29
M3 Ultra · 96 GB	Q4_K_M	Runs great	45 GB	55
M3 Ultra · 256 GB	Q8_0	Runs great	78 GB	47
M3 Ultra · 512 GB	Q8_0	Runs great	78 GB	47
M2 · 8 GB	Q3_K_L	Too big	38 GB
M2 · 16 GB	Q3_K_L	Too big	38 GB
M2 · 24 GB	Q3_K_L	Too big	38 GB
M2 Pro · 16 GB	Q3_K_L	Too big	38 GB
M2 Pro · 32 GB	Q3_K_L	Too big	38 GB
M2 Max · 32 GB	Q3_K_L	Too big	38 GB
M2 Max · 64 GB	Q4_K_M	Runs well	42 GB	35
M2 Max · 96 GB	Q4_K_M	Runs great	45 GB	35
M2 Ultra · 64 GB	Q4_K_M	Runs well	42 GB	54
M2 Ultra · 128 GB	Q6_K	Runs great	61 GB	47
M2 Ultra · 192 GB	Q8_0	Runs great	78 GB	47
M1 · 8 GB	Q3_K_L	Too big	38 GB
M1 · 16 GB	Q3_K_L	Too big	38 GB
M1 Pro · 16 GB	Q3_K_L	Too big	38 GB
M1 Pro · 32 GB	Q3_K_L	Too big	38 GB
M1 Max · 32 GB	Q3_K_L	Too big	38 GB
M1 Max · 64 GB	Q4_K_M	Runs well	42 GB	29
M1 Ultra · 64 GB	Q4_K_M	Runs well	42 GB	42
M1 Ultra · 128 GB	Q6_K	Runs great	61 GB	38

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

Information

Maker: meituan-longcat
Released: January 2026
Parameters: 69B, 4.1B active
Longest chat: 320K tokens
Reads pictures: No
Licence: mit
Downloads this month: 3,623
Likes: 10
Sizes: 6, 33 to 137 GB
Kind: Maker's release
Source: meituan-longcat/LongCat-Flash-Lite

Badge for the model card

Shows on LongCat Flash Lite'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.