Hugging FaceA community version of another maker's model that reads pictures as well as text.
- Rating
- 3.8
- Downloads
- 54K
- Likes
- 45
- Parameters
- 107B, 13B active
- Context
- 128K
- Licence
- mit
Too big for this Mac
Its smallest size needs 48 GB of memory, more than M4 · 16 GB can give it.
Ratings
3.8 out of 5, across people's blind votes on its answers.
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.
GLM 4.6V runs well from a Mac with an M1 Max and 64 GB of memory. 24 of 59 Mac configurations run it well. It is a re-tune or merge made by someone other than the original model's maker, so its ratings, where it has any, are its own.
Compare with
The nearest models in size.
- Qwen3.8 Flash Next Avalonfreriadoc · 5.7K downloadsToo big for this Mac · 95 GB
- C4ai Command R Plus 08 2024Cohere · 8.9K downloadsToo big for this Mac · 23 GB
- Ling Flash 2.0Ant Group · 4.8K downloads · Arena 1343Too big for this Mac · 21 GB
- Qwen1.5 110BAlibaba Qwen · 5.4K downloads · Arena 1234Too big for this Mac · 24 GB
Sizes
From unsloth/GLM-4.6V-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.
| Size | Download | Memory at 8K | Get |
|---|---|---|---|
| UD-TQ1_0 | 37 GB | 38 GBestimated | Open |
| UD-IQ1_S | 37 GB | 38 GBestimated | Open |
| UD-IQ1_M | 39 GB | 40 GBestimated | Open |
| UD-IQ2_XXS | 41 GB | 42 GBestimated | Open |
| UD-IQ2_M | 43 GB | 44 GBestimated | Open |
| Q2_K | 44 GB | 45 GBestimated | Open |
| Q2_K_L | 44 GB | 45 GBestimated | Open |
| UD-Q2_K_XL | 46 GB | 47 GBestimated | Open |
| UD-IQ3_XXS | 50 GB | 51 GBestimated | Open |
| Q3_K_S | 51 GB | 52 GBestimated | Open |
| UD-Q3_K_XL | 53 GB | 54 GBestimated | Open |
| Q3_K_M | 55 GB | 56 GBestimated | Open |
| IQ4_XS | 58 GB | 59 GBestimated | Open |
| IQ4_NL | 61 GB | 62 GBestimated | Open |
| Q4_0 | 61 GB | 62 GBestimated | Open |
| Q4_K_S | 65 GB | 66 GBestimated | Open |
| UD-Q4_K_XL | 65 GB | 66 GBestimated | Open |
| Q4_1 | 67 GB | 68 GBestimated | Open |
| Q4_K_M | 71 GB | 72 GBestimated | Open |
| Q5_K_S | 76 GB | 77 GBestimated | Open |
| UD-Q5_K_XL | 80 GB | 81 GBestimated | Open |
| Q5_K_M | 81 GB | 82 GBestimated | Open |
| Q6_K | 96 GB | 97 GBestimated | Open |
| UD-Q6_K_XL | 98 GB | 99 GBestimated | Open |
| Q8_0 | 114 GB | 115 GBestimated | Open |
| UD-Q8_K_XL | 124 GB | 125 GBestimated | Open |
| BF16 | 214 GB | 215 GBestimated | Open |
Run It Anywhere
GLM 4.6V is an open model, so any app that runs GGUF files can run it. These take the UD-TQ1_0 size from unsloth straight from Hugging Face.
- LM StudioMac, Windows, Linux
Open - JanMac, Windows, Linux
Open - MstyMac, Windows, Linux
Open - OllamaMac, Windows, Linux
ollama run hf.co/unsloth/GLM-4.6V-GGUF:UD-TQ1_0 - llama.cppMac, Windows, Linux
llama-server -hf unsloth/GLM-4.6V-GGUF:UD-TQ1_0 - Docker Model RunnerMac, Windows, Linux
docker model run hf.co/unsloth/GLM-4.6V-GGUF:UD-TQ1_0
On a PC Graphics Card
| Video memory | Largest size that fits | Download | Memory at 8K |
|---|---|---|---|
| 6 GB | None worth running | needs 44 GB | |
| 8 GB | None worth running | needs 44 GB | |
| 10 GB | None worth running | needs 44 GB | |
| 11 GB | None worth running | needs 44 GB | |
| 12 GB | None worth running | needs 44 GB | |
| 16 GB | None worth running | needs 44 GB | |
| 20 GB | None worth running | needs 44 GB | |
| 24 GB | None worth running | needs 44 GB | |
| 32 GB | None worth running | needs 44 GB | |
| 48 GB | UD-Q2_K_XL | 46 GB | 47 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 | 48 GB | |
| M6 · 24 GB | IQ1_S | Too big | 48 GB | |
| M6 · 32 GB | IQ1_S | Too big | 48 GB | |
| M5 · 16 GB | IQ1_S | Too big | 48 GB | |
| M5 · 24 GB | IQ1_S | Too big | 48 GB | |
| M5 · 32 GB | IQ1_S | Too big | 48 GB | |
| M5 Pro · 24 GB | IQ1_S | Too big | 48 GB | |
| M5 Pro · 48 GB | UD-IQ1_S | Tight fit | 38 GB | 11 |
| M5 Pro · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 11 |
| M5 Max · 36 GB | IQ1_S | Too big | 48 GB | |
| M5 Max · 48 GB | UD-IQ1_S | Tight fit | 38 GB | 21 |
| M5 Max · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 20 |
| M5 Max · 128 GB | Q4_1 | Runs great | 68 GB | 18 |
| M5 Ultra · 96 GB | Q4_1 | Runs well | 68 GB | 34 |
| M5 Ultra · 256 GB | Q8_0 | Runs great | 128 GB | 26 |
| M5 Ultra · 512 GB | Q8_0 | Runs great | 128 GB | 26 |
| M4 · 16 GB | IQ1_S | Too big | 48 GB | |
| M4 · 24 GB | IQ1_S | Too big | 48 GB | |
| M4 · 32 GB | IQ1_S | Too big | 48 GB | |
| M4 Pro · 24 GB | IQ1_S | Too big | 48 GB | |
| M4 Pro · 48 GB | UD-IQ1_S | Tight fit | 38 GB | 9.1 |
| M4 Pro · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 8.7 |
| M4 Max · 36 GB | IQ1_S | Too big | 48 GB | |
| M4 Max · 48 GB | UD-IQ1_S | Tight fit | 38 GB | 16 |
| M4 Max · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 15 |
| M4 Max · 96 GB | Q4_1 | Runs well | 68 GB | 11 |
| M4 Max · 128 GB | Q4_1 | Runs great | 68 GB | 14 |
| M3 · 8 GB | IQ1_S | Too big | 48 GB | |
| M3 · 16 GB | IQ1_S | Too big | 48 GB | |
| M3 · 24 GB | IQ1_S | Too big | 48 GB | |
| M3 Pro · 18 GB | IQ1_S | Too big | 48 GB | |
| M3 Pro · 36 GB | IQ1_S | Too big | 48 GB | |
| M3 Max · 36 GB | IQ1_S | Too big | 48 GB | |
| M3 Max · 48 GB | UD-IQ1_S | Tight fit | 38 GB | 13 |
| M3 Max · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 12 |
| M3 Max · 96 GB | Q4_1 | Runs well | 68 GB | 8.8 |
| M3 Max · 128 GB | Q4_1 | Runs great | 68 GB | 11 |
| M3 Ultra · 96 GB | Q4_1 | Runs well | 68 GB | 20 |
| M3 Ultra · 256 GB | Q8_0 | Runs great | 128 GB | 15 |
| M3 Ultra · 512 GB | Q8_0 | Runs great | 128 GB | 15 |
| M2 · 8 GB | IQ1_S | Too big | 48 GB | |
| M2 · 16 GB | IQ1_S | Too big | 48 GB | |
| M2 · 24 GB | IQ1_S | Too big | 48 GB | |
| M2 Pro · 16 GB | IQ1_S | Too big | 48 GB | |
| M2 Pro · 32 GB | IQ1_S | Too big | 48 GB | |
| M2 Max · 32 GB | IQ1_S | Too big | 48 GB | |
| M2 Max · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 12 |
| M2 Max · 96 GB | Q4_1 | Runs well | 68 GB | 11 |
| M2 Ultra · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 21 |
| M2 Ultra · 128 GB | Q4_1 | Runs great | 68 GB | 20 |
| M2 Ultra · 192 GB | Q8_0 | Runs great | 128 GB | 15 |
| M1 · 8 GB | IQ1_S | Too big | 48 GB | |
| M1 · 16 GB | IQ1_S | Too big | 48 GB | |
| M1 Pro · 16 GB | IQ1_S | Too big | 48 GB | |
| M1 Pro · 32 GB | IQ1_S | Too big | 48 GB | |
| M1 Max · 32 GB | IQ1_S | Too big | 48 GB | |
| M1 Max · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 11 |
| M1 Ultra · 64 GB | UD-IQ2_XXS | Runs well | 42 GB | 18 |
| M1 Ultra · 128 GB | Q4_1 | Runs great | 68 GB | 17 |
Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.
Packaged by
3 repositories on Hugging Face carry this model for llama.cpp. Outmake reads them all as one model.
Information
- Maker
- Z.ai
- Released
- December 2025
- Parameters
- 107B, 13B active
- Longest chat
- 128K tokens
- Reads pictures
- Yes
- Licence
- mit
- Downloads this month
- 54,374
- Likes
- 45
- Sizes
- 27, 37 to 214 GB
- Kind
- Community version
- Source
- zai-org/GLM-4.6V
Badge for the model card
Shows on GLM 4.6V'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.