Outmake Models
Models

Hy MT2 7B

Tencent · May 2026

Get Hugging Face

A 7.5B chat model from Tencent.

Downloads
1.1M
Likes
99
Parameters
7.5B
Context
256K
Licence
apache-2.0

Runs great on M4 · 16 GB

The Q6_K size from tencent is the one to get: a 6.2 GB download that uses 7.5 GB of memory, and writes about 6.5 words a second.

Hy MT2 7B runs well from a Mac with an M1 and 16 GB of memory. 56 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From tencent/Hy-MT2-7B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q4_K_M4.6 GB5.4 GBestimatedOpen
Q6_K6.2 GB6.9 GBestimatedOpen
Q8_08.0 GB8.7 GBestimatedOpen

Run It Anywhere

Hy MT2 7B is an open model, so any app that runs GGUF files can run it. These take the Q6_K size from tencent straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/tencent/Hy-MT2-7B-GGUF:Q6_K
  • llama.cppMac, Windows, Linuxllama-server -hf tencent/Hy-MT2-7B-GGUF:Q6_K
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/tencent/Hy-MT2-7B-GGUF:Q6_K
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBQ4_K_M4.6 GB5.4 GB
8 GBQ6_K6.2 GB6.9 GB
10 GBQ8_08.0 GB8.7 GB
11 GBQ8_08.0 GB8.7 GB
12 GBQ8_08.0 GB8.7 GB
16 GBQ8_08.0 GB8.7 GB
20 GBQ8_08.0 GB8.7 GB
24 GBQ8_08.0 GB8.7 GB
32 GBQ8_08.0 GB8.7 GB
48 GBQ8_08.0 GB8.7 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 great7.5 GB9.6
M6 · 24 GBQ8_0Runs great11 GB9.3
M6 · 32 GBQ8_0Runs great13 GB9.3
M5 · 16 GBQ6_KRuns great7.5 GB8.7
M5 · 24 GBQ8_0Runs great11 GB8.4
M5 · 32 GBQ8_0Runs great13 GB8.4
M5 Pro · 24 GBQ8_0Runs great11 GB17
M5 Pro · 48 GBQ8_0Runs great18 GB17
M5 Pro · 64 GBQ8_0Runs great28 GB17
M5 Max · 36 GBQ8_0Runs great13 GB25
M5 Max · 48 GBQ8_0Runs great18 GB32
M5 Max · 64 GBQ8_0Runs great28 GB32
M5 Max · 128 GBQ8_0Runs great28 GB32
M5 Ultra · 96 GBQ8_0Runs great28 GB61
M5 Ultra · 256 GBQ8_0Runs great28 GB61
M5 Ultra · 512 GBQ8_0Runs great28 GB61
M4 · 16 GBQ6_KRuns great7.5 GB6.5
M4 · 24 GBQ8_0Runs great11 GB6.4
M4 · 32 GBQ8_0Runs great13 GB6.4
M4 Pro · 24 GBQ8_0Runs great11 GB14
M4 Pro · 48 GBQ8_0Runs great18 GB14
M4 Pro · 64 GBQ8_0Runs great28 GB14
M4 Max · 36 GBQ8_0Runs great13 GB20
M4 Max · 48 GBQ8_0Runs great18 GB25
M4 Max · 64 GBQ8_0Runs great28 GB25
M4 Max · 96 GBQ8_0Runs great28 GB20
M4 Max · 128 GBQ8_0Runs great28 GB25
M3 · 8 GBQ4_K_MTight fit5.4 GB6.5
M3 · 16 GBQ6_KRuns great7.5 GB5.5
M3 · 24 GBQ8_0Runs great11 GB5.4
M3 Pro · 18 GBQ6_KRuns great7.5 GB8.2
M3 Pro · 36 GBQ8_0Runs great13 GB8
M3 Max · 36 GBQ8_0Runs great13 GB15
M3 Max · 48 GBQ8_0Runs great18 GB20
M3 Max · 64 GBQ8_0Runs great28 GB20
M3 Max · 96 GBQ8_0Runs great28 GB15
M3 Max · 128 GBQ8_0Runs great28 GB20
M3 Ultra · 96 GBQ8_0Runs great28 GB36
M3 Ultra · 256 GBQ8_0Runs great28 GB36
M3 Ultra · 512 GBQ8_0Runs great28 GB36
M2 · 8 GBQ4_K_MTight fit5.4 GB6.4
M2 · 16 GBQ6_KRuns great7.5 GB5.5
M2 · 24 GBQ8_0Runs great11 GB5.4
M2 Pro · 16 GBQ6_KRuns great7.5 GB11
M2 Pro · 32 GBQ8_0Runs great13 GB10
M2 Max · 32 GBQ8_0Runs great13 GB20
M2 Max · 64 GBQ8_0Runs great28 GB20
M2 Max · 96 GBQ8_0Runs great28 GB20
M2 Ultra · 64 GBQ8_0Runs great28 GB35
M2 Ultra · 128 GBQ8_0Runs great28 GB35
M2 Ultra · 192 GBQ8_0Runs great28 GB35
M1 · 8 GBQ4_K_MTight fit5.4 GB4.4
M1 · 16 GBQ6_KRuns great7.5 GB3.7
M1 Pro · 16 GBQ6_KRuns great7.5 GB10
M1 Pro · 32 GBQ8_0Runs great13 GB10
M1 Max · 32 GBQ8_0Runs great13 GB18
M1 Max · 64 GBQ8_0Runs great28 GB18
M1 Ultra · 64 GBQ8_0Runs great28 GB31
M1 Ultra · 128 GBQ8_0Runs great28 GB31

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

Information

Maker
Tencent
Released
May 2026
Parameters
7.5B
Longest chat
256K tokens
Reads pictures
No
Licence
apache-2.0
Downloads this month
1,094,212
Likes
99
Sizes
3, 4.6 to 8.0 GB
Kind
Maker's release

Badge for the model card

Shows on Hy MT2 7B'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