Outmake Models
Models

GLM 4.7 Flash

Z.ai · January 2026

GetHugging Face

GLM 4.7 Flash is meant for lightweight use, balancing performance against efficiency.

Rating
3.6
Downloads
290K
Parameters
31B, 3B active
Context
198K
Download
8.3 to 60 GB
Commercial use
Yes

Tight fit on M4 · 16 GB

Outmake picks the UD-Q2_K_XL size: a 12 GB download that uses 12 GB of memory, and writes about 14 words (19 tokens) a second.

About

Only about a tenth of the model is in use at any one time.

How it compares

Of the 8 chat models Outmake recommends for a 24 GB Mac, it is the 4th fastest by Outmake's estimate.

Outmake estimates Qwen3.6 35B writes about 1.6 times as fast, and Qwen3.5 4B at about the same speed.

Of the 3 of those 8 that LM Arena rates, it places 3rd in the public vote, behind Qwen3.8 27B and Gemma 4 26B.

Good to know

GLM-4.5, the model in its cited report, was trained in several stages, then refined with trial-and-reward training.

The GLM family also includes GLM-4.5 and a compact version, GLM-4.5-Air.

Sources: zai-org/GLM-4.7-Flash model card, GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models, GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models, Outmake's speed estimates, LM Arena and Hugging Face. Every line comes from one of these.

Ratings

3.6 out of 5, from how people voted on its answers.

Text
1365
Coding
1423

LM Arena ratings of October 28, 2026 (CC BY 4.0): people vote between two answers without knowing which model wrote which. The bar ends at 1488, the best any open model has. The Arena tests the full model; a small size on a Mac answers a little worse.

Good at

  • Thinking before it answers

Licence

MIT. Commercial use: yes. The maker's page has the full terms.

GLM 4.7 Flash runs well from a Mac with an M3 Pro and 18 GB of memory, at Q2_K_L, writing about 17 words (23 tokens) a second. 48 of 59 Mac configurations run it well.

Compare with

The nearest models in size, judged for the same Mac.

Sizes

SizeDownloadMemory at 8KMemory at 32KOpen
IQ4_XSSmaller16 GB17 GB17 GBOpen
UD-Q4_K_XLBalanced18 GB18 GB19 GBOpen
Q6_KBest25 GB25 GB26 GBOpen
All 28 sizes
SizeDownloadMemory at 8KMemory at 32KOpen
UD-TQ1_08.3 GB8.7 GBestimated9.4 GBOpen
UD-IQ1_S9.2 GB9.6 GBestimated10 GBOpen
UD-IQ1_M9.8 GB10 GBestimated11 GBOpen
UD-IQ2_XXS11 GB11 GBestimated12 GBOpen
UD-IQ2_M11 GB11 GBestimated12 GBOpen
Q2_K11 GB12 GBestimated12 GBOpen
Q2_K_L11 GB12 GBestimated12 GBOpen
UD-Q2_K_XL12 GB12 GBestimated13 GBOpen
UD-IQ3_XXS13 GB13 GBestimated14 GBOpen
Q3_K_S13 GB14 GBestimated14 GBOpen
UD-Q3_K_XL14 GB14 GBestimated15 GBOpen
Q3_K_M15 GB15 GBestimated16 GBOpen
MXFP4_MOE17 GB17 GBestimated18 GBOpen
IQ4_NL17 GB18 GBestimated18 GBOpen
Q4_017 GB18 GBestimated18 GBOpen
Q4_K_S17 GB18 GBestimated18 GBOpen
Q4_K_M18 GB19 GBestimated19 GBOpen
Q4_119 GB19 GBestimated20 GBOpen
Q5_K_S21 GB21 GBestimated22 GBOpen
Q5_K_M21 GB22 GBestimated22 GBOpen
UD-Q5_K_XL22 GB22 GBestimated23 GBOpen
UD-Q6_K_XL26 GB26 GBestimated27 GBOpen
Q8_032 GB32 GBestimated33 GBOpen
UD-Q8_K_XL36 GB36 GBestimated37 GBOpen
BF1660 GB60 GBestimated61 GBOpen

Memory is what llama.cpp's own calculator reported for each size when the catalog was built, with the app's settings. Smaller, Balanced and Best are the three sizes the store names.

Run It Anywhere

GLM 4.7 Flash is an open model, so any app that runs GGUF files can run it. These take the UD-Q4_K_XL size from unsloth straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/unsloth/GLM-4.7-Flash-GGUF:UD-Q4_K_XL
  • llama.cppMac, Windows, Linuxllama-server -hf unsloth/GLM-4.7-Flash-GGUF:UD-Q4_K_XL
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/unsloth/GLM-4.7-Flash-GGUF:UD-Q4_K_XL
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 11 GB
8 GBNone worth runningneeds 11 GB
10 GBNone worth runningneeds 11 GB
11 GBNone worth runningneeds 11 GB
12 GBUD-IQ2_M11 GB11 GB
16 GBQ3_K_M15 GB15 GB
20 GBQ4_119 GB19 GB
24 GBUD-Q5_K_XL22 GB22 GB
32 GBUD-Q6_K_XL26 GB26 GB
48 GBUD-Q8_K_XL36 GB36 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
MacSizeFitMemoryWritesReads
M6 · 16 GBUD-Q2_K_XLTight fit12 GB221467
M6 · 24 GBQ3_K_MRuns well16 GB201467
M6 · 32 GBUD-Q4_K_XLRuns well19 GB191467
M5 · 16 GBUD-Q2_K_XLTight fit12 GB21542
M5 · 24 GBQ3_K_MRuns well16 GB19542
M5 · 32 GBUD-Q4_K_XLRuns well19 GB17542
M5 Pro · 24 GBQ3_K_MRuns well16 GB351083
M5 Pro · 48 GBUD-Q4_K_XLRuns great22 GB321083
M5 Pro · 64 GBQ6_KRuns great29 GB281354
M5 Max · 36 GBIQ4_XSRuns great17 GB472167
M5 Max · 48 GBUD-Q4_K_XLRuns great22 GB572709
M5 Max · 64 GBQ6_KRuns great29 GB512709
M5 Max · 128 GBQ6_KRuns great29 GB512709
M5 Ultra · 96 GBQ6_KRuns great29 GB844334
M5 Ultra · 256 GBQ6_KRuns great29 GB845417
M5 Ultra · 512 GBQ6_KRuns great29 GB845417
M4 · 16 GBUD-Q2_K_XLTight fit12 GB14182
M4 · 24 GBQ3_K_MRuns well16 GB13182
M4 · 32 GBUD-Q4_K_XLRuns well19 GB12182
M4 Pro · 24 GBQ3_K_MRuns well16 GB23364
M4 Pro · 48 GBUD-Q4_K_XLRuns great22 GB22364
M4 Pro · 64 GBQ6_KRuns great29 GB20364
M4 Max · 36 GBIQ4_XSRuns great17 GB29728
M4 Max · 48 GBUD-Q4_K_XLRuns great22 GB33910
M4 Max · 64 GBQ6_KRuns great29 GB30910
M4 Max · 96 GBQ6_KRuns great29 GB25728
M4 Max · 128 GBQ6_KRuns great29 GB30910
M3 · 8 GBIQ4_XSToo big20 GB
M3 · 16 GBUD-Q2_K_XLTight fit12 GB12151
M3 · 24 GBQ3_K_MRuns well16 GB12151
M3 Pro · 18 GBQ2_K_LRuns well12 GB17264
M3 Pro · 36 GBIQ4_XSRuns great17 GB15264
M3 Max · 36 GBIQ4_XSRuns great17 GB26565
M3 Max · 48 GBUD-Q4_K_XLRuns great22 GB30753
M3 Max · 64 GBQ6_KRuns great29 GB27753
M3 Max · 96 GBQ6_KRuns great29 GB22565
M3 Max · 128 GBQ6_KRuns great29 GB27753
M3 Ultra · 96 GBQ6_KRuns great29 GB391130
M3 Ultra · 256 GBQ6_KRuns great29 GB391130
M3 Ultra · 512 GBQ6_KRuns great29 GB391507
M2 · 8 GBIQ4_XSToo big20 GB
M2 · 16 GBUD-Q2_K_XLTight fit12 GB12148
M2 · 24 GBQ3_K_MRuns well16 GB11148
M2 Pro · 16 GBUD-Q2_K_XLTight fit12 GB21296
M2 Pro · 32 GBUD-Q4_K_XLRuns well19 GB19296
M2 Max · 32 GBUD-Q4_K_XLRuns well19 GB30555
M2 Max · 64 GBQ6_KRuns great29 GB26555
M2 Max · 96 GBQ6_KRuns great29 GB26703
M2 Ultra · 64 GBQ6_KRuns great29 GB391109
M2 Ultra · 128 GBQ6_KRuns great29 GB391109
M2 Ultra · 192 GBQ6_KRuns great29 GB391109
M1 · 8 GBIQ4_XSToo big20 GB
M1 · 16 GBUD-Q2_K_XLTight fit12 GB8.2120
M1 Pro · 16 GBUD-Q2_K_XLTight fit12 GB18240
M1 Pro · 32 GBUD-Q4_K_XLRuns well19 GB16240
M1 Max · 32 GBUD-Q4_K_XLRuns well19 GB23411
M1 Max · 64 GBQ6_KRuns great29 GB21411
M1 Ultra · 64 GBQ6_KRuns great29 GB29822
M1 Ultra · 128 GBQ6_KRuns great29 GB29822

Writes and reads are words a second (a token is about three quarters of a word), 4K tokens into a chat, estimated unless marked measured. The size is the one the store offers that Mac. Constants fitted October 11, 2026; how, and how wrong.

Badge for the model card

Shows on GLM 4.7 Flash's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Runs on Mac badge

Sizes and files are read from unsloth/GLM-4.7-Flash-GGUF on Hugging Face and checked every morning. The rating is the model's score on the LM Arena text leaderboard (CC BY 4.0), where people vote between two answers without knowing which model wrote which. Catalog of November 4, 2026.

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