Hugging FaceA community version of another maker's model that reads pictures as well as text.
- Rating
- 4.0
- Downloads
- 2.4K
- Likes
- 52
- Parameters
- 235B, 22B active
- Context
- 256K
- Licence
- apache-2.0
Too big for this Mac
Its smallest size needs 171 GB of memory, more than M4 · 16 GB can give it.
Ratings
4.0 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.
Qwen3 VL 235B A22B Thinking runs well from a Mac with an M2 Ultra and 192 GB of memory. 5 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 235B A22B 2507Alibaba Qwen · 24K downloadsToo big for this Mac · 53 GB
- Qwen3 235B A22BAlibaba Qwen · 18K downloads · Arena 1375Too big for this Mac · 86 GB
- MiniMax M2.7MiniMax · 15K downloads · Arena 1415Too big for this Mac · 47 GB
- MiniMax M2MiniMax · 8.8K downloads · Arena 1344Too big for this Mac · 56 GB
Sizes
From Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.
| Size | Download | Memory at 8K | Get |
|---|---|---|---|
| Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | 142 GB | 143 GBestimated | Open |
| Q8_0 | 250 GB | 251 GBestimated | Open |
| Qwen3VL-235B-A22B-Thinking-F16-split | 470 GB | 471 GBestimated | Open |
Run It Anywhere
Qwen3 VL 235B A22B Thinking is an open model, so any app that runs GGUF files can run it. These take the Qwen3VL-235B-A22B-Thinking-Q4_K_M-split size from Qwen straight from Hugging Face.
- LM StudioMac, Windows, Linux
Open - JanMac, Windows, Linux
Open - MstyMac, Windows, Linux
Open - OllamaMac, Windows, Linux
ollama run hf.co/Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF:Qwen3VL-235B-A22B-Thinking-Q4_K_M-split - llama.cppMac, Windows, Linux
llama-server -hf Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF:Qwen3VL-235B-A22B-Thinking-Q4_K_M-split - Docker Model RunnerMac, Windows, Linux
docker model run hf.co/Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF:Qwen3VL-235B-A22B-Thinking-Q4_K_M-split
On a PC Graphics Card
| Video memory | Largest size that fits | Download | Memory at 8K |
|---|---|---|---|
| 6 GB | None worth running | needs 143 GB | |
| 8 GB | None worth running | needs 143 GB | |
| 10 GB | None worth running | needs 143 GB | |
| 11 GB | None worth running | needs 143 GB | |
| 12 GB | None worth running | needs 143 GB | |
| 16 GB | None worth running | needs 143 GB | |
| 20 GB | None worth running | needs 143 GB | |
| 24 GB | None worth running | needs 143 GB | |
| 32 GB | None worth running | needs 143 GB | |
| 48 GB | None worth running | needs 143 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 | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M6 · 24 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M6 · 32 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 · 16 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 · 24 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 · 32 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Pro · 24 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Pro · 48 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Pro · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Max · 36 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Max · 48 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Max · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Max · 128 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Ultra · 96 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M5 Ultra · 256 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Runs great | 171 GB | 21 |
| M5 Ultra · 512 GB | Q8_0 | Runs great | 279 GB | 17 |
| M4 · 16 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 · 24 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 · 32 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Pro · 24 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Pro · 48 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Pro · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Max · 36 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Max · 48 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Max · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Max · 96 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M4 Max · 128 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 · 8 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 · 16 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 · 24 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Pro · 18 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Pro · 36 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Max · 36 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Max · 48 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Max · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Max · 96 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Max · 128 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Ultra · 96 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M3 Ultra · 256 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Runs great | 171 GB | 12 |
| M3 Ultra · 512 GB | Q8_0 | Runs great | 279 GB | 9.8 |
| M2 · 8 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 · 16 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 · 24 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Pro · 16 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Pro · 32 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Max · 32 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Max · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Max · 96 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Ultra · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Ultra · 128 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M2 Ultra · 192 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Runs well | 150 GB | 12 |
| M1 · 8 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M1 · 16 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M1 Pro · 16 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M1 Pro · 32 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M1 Max · 32 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M1 Max · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M1 Ultra · 64 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 GB | |
| M1 Ultra · 128 GB | Qwen3VL-235B-A22B-Thinking-Q4_K_M-split | Too big | 171 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
- Alibaba Qwen
- Released
- September 2025
- Parameters
- 235B, 22B active
- Longest chat
- 256K tokens
- Reads pictures
- Yes
- Licence
- apache-2.0
- Downloads this month
- 2,449
- Likes
- 52
- Sizes
- 3, 142 to 470 GB
- Kind
- Community version
Badge for the model card
Shows on Qwen3 VL 235B A22B Thinking'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.