Hugging FaceA 33B chat model from LG AI that reads pictures as well as text.
- Rating
- 4.4
- Downloads
- 4.7K
- Likes
- 55
- Parameters
- 33B
- Context
- 256K
- Licence
- other
Too big for this Mac
Its smallest size needs 28 GB of memory, more than M4 · 16 GB can give it.
Ratings
4.4 out of 5, across the tests below.
MMLU-Pro: graduate-level questions in 14 subjects. GPQA Diamond: questions PhD scientists write for each other. AIME 2026: this year's AIME competition problems. 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.
EXAONE 4.5 33B runs well from a Mac with an M1 Pro and 32 GB of memory. 40 of 59 Mac configurations run it well.
Compare with
The nearest models in size.
- Nanbeige4.3 33BNanbeige · 204K downloadsRuns well · 8.3 GB · 4.5 words/s
- NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA · 2.4K downloadsToo big for this Mac · 19 GB
- WhiteRabbitNeo 33B V1.5manuelgutierrez · 70K downloadsToo big for this Mac · 19 GB
- Qwen2.5 32BAlibaba Qwen · 833K downloadsRuns well · 9.0 GB · 3.8 words/s
Sizes
From LGAI-EXAONE/EXAONE-4.5-33B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.
| Size | Download | Memory at 8K | Get |
|---|---|---|---|
| IQ4_XS | 18 GB | 19 GBestimated | Open |
| Q4_K_M | 20 GB | 21 GBestimated | Open |
| Q5_K_M | 23 GB | 24 GBestimated | Open |
| Q6_K | 27 GB | 28 GBestimated | Open |
| Q8_0 | 35 GB | 36 GBestimated | Open |
| BF16 | 66 GB | 67 GBestimated | Open |
Run It Anywhere
EXAONE 4.5 33B is an open model, so any app that runs GGUF files can run it. These take the IQ4_XS size from LGAI-EXAONE straight from Hugging Face.
- LM StudioMac, Windows, Linux
Open - JanMac, Windows, Linux
Open - MstyMac, Windows, Linux
Open - OllamaMac, Windows, Linux
ollama run hf.co/LGAI-EXAONE/EXAONE-4.5-33B-GGUF:IQ4_XS - llama.cppMac, Windows, Linux
llama-server -hf LGAI-EXAONE/EXAONE-4.5-33B-GGUF:IQ4_XS - Docker Model RunnerMac, Windows, Linux
docker model run hf.co/LGAI-EXAONE/EXAONE-4.5-33B-GGUF:IQ4_XS
On a PC Graphics Card
| Video memory | Largest size that fits | Download | Memory at 8K |
|---|---|---|---|
| 6 GB | None worth running | needs 19 GB | |
| 8 GB | None worth running | needs 19 GB | |
| 10 GB | None worth running | needs 19 GB | |
| 11 GB | None worth running | needs 19 GB | |
| 12 GB | None worth running | needs 19 GB | |
| 16 GB | None worth running | needs 19 GB | |
| 20 GB | IQ4_XS | 18 GB | 19 GB |
| 24 GB | Q4_K_M | 20 GB | 21 GB |
| 32 GB | Q6_K | 27 GB | 28 GB |
| 48 GB | Q8_0 | 35 GB | 36 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 | IQ4_XS | Too big | 28 GB | |
| M6 · 24 GB | IQ4_XS | Too big | 28 GB | |
| M6 · 32 GB | Q4_K_M | Runs well | 21 GB | 4.3 |
| M5 · 16 GB | IQ4_XS | Too big | 28 GB | |
| M5 · 24 GB | IQ4_XS | Too big | 28 GB | |
| M5 · 32 GB | Q4_K_M | Runs well | 21 GB | 3.9 |
| M5 Pro · 24 GB | IQ4_XS | Too big | 28 GB | |
| M5 Pro · 48 GB | Q4_K_M | Runs great | 23 GB | 7.6 |
| M5 Pro · 64 GB | Q6_K | Runs great | 30 GB | 6.2 |
| M5 Max · 36 GB | Q4_K_M | Runs well | 23 GB | 11 |
| M5 Max · 48 GB | Q4_K_M | Runs great | 23 GB | 15 |
| M5 Max · 64 GB | Q6_K | Runs great | 30 GB | 12 |
| M5 Max · 128 GB | Q8_0 | Runs great | 46 GB | 11 |
| M5 Ultra · 96 GB | Q8_0 | Runs great | 46 GB | 21 |
| M5 Ultra · 256 GB | Q8_0 | Runs great | 46 GB | 21 |
| M5 Ultra · 512 GB | Q8_0 | Runs great | 46 GB | 21 |
| M4 · 16 GB | IQ4_XS | Too big | 28 GB | |
| M4 · 24 GB | IQ4_XS | Too big | 28 GB | |
| M4 · 32 GB | Q4_K_M | Runs well | 21 GB | 3.1 |
| M4 Pro · 24 GB | IQ4_XS | Too big | 28 GB | |
| M4 Pro · 48 GB | Q4_K_M | Runs great | 23 GB | 6.6 |
| M4 Pro · 64 GB | Q6_K | Runs great | 30 GB | 5.1 |
| M4 Max · 36 GB | Q4_K_M | Runs well | 23 GB | 9.4 |
| M4 Max · 48 GB | Q4_K_M | Runs great | 23 GB | 12 |
| M4 Max · 64 GB | Q6_K | Runs great | 30 GB | 9.5 |
| M4 Max · 96 GB | Q8_0 | Runs great | 46 GB | 6.8 |
| M4 Max · 128 GB | Q8_0 | Runs great | 46 GB | 8.7 |
| M3 · 8 GB | IQ4_XS | Too big | 28 GB | |
| M3 · 16 GB | IQ4_XS | Too big | 28 GB | |
| M3 · 24 GB | IQ4_XS | Too big | 28 GB | |
| M3 Pro · 18 GB | IQ4_XS | Too big | 28 GB | |
| M3 Pro · 36 GB | Q4_K_M | Runs well | 23 GB | 3.8 |
| M3 Max · 36 GB | Q4_K_M | Runs well | 23 GB | 7.3 |
| M3 Max · 48 GB | Q4_K_M | Runs great | 23 GB | 9.4 |
| M3 Max · 64 GB | Q6_K | Runs great | 30 GB | 7.4 |
| M3 Max · 96 GB | Q8_0 | Runs great | 46 GB | 5.1 |
| M3 Max · 128 GB | Q8_0 | Runs great | 46 GB | 6.7 |
| M3 Ultra · 96 GB | Q8_0 | Runs great | 46 GB | 12 |
| M3 Ultra · 256 GB | Q8_0 | Runs great | 46 GB | 12 |
| M3 Ultra · 512 GB | Q8_0 | Runs great | 46 GB | 12 |
| M2 · 8 GB | IQ4_XS | Too big | 28 GB | |
| M2 · 16 GB | IQ4_XS | Too big | 28 GB | |
| M2 · 24 GB | IQ4_XS | Too big | 28 GB | |
| M2 Pro · 16 GB | IQ4_XS | Too big | 28 GB | |
| M2 Pro · 32 GB | Q4_K_M | Runs well | 21 GB | 5 |
| M2 Max · 32 GB | Q4_K_M | Runs well | 21 GB | 9.4 |
| M2 Max · 64 GB | Q6_K | Runs great | 30 GB | 7.4 |
| M2 Max · 96 GB | Q8_0 | Runs great | 46 GB | 6.7 |
| M2 Ultra · 64 GB | Q6_K | Runs great | 30 GB | 13 |
| M2 Ultra · 128 GB | Q8_0 | Runs great | 46 GB | 12 |
| M2 Ultra · 192 GB | Q8_0 | Runs great | 46 GB | 12 |
| M1 · 8 GB | IQ4_XS | Too big | 28 GB | |
| M1 · 16 GB | IQ4_XS | Too big | 28 GB | |
| M1 Pro · 16 GB | IQ4_XS | Too big | 28 GB | |
| M1 Pro · 32 GB | Q4_K_M | Runs well | 21 GB | 4.8 |
| M1 Max · 32 GB | Q4_K_M | Runs well | 21 GB | 8.7 |
| M1 Max · 64 GB | Q6_K | Runs great | 30 GB | 6.9 |
| M1 Ultra · 64 GB | Q6_K | Runs great | 30 GB | 12 |
| M1 Ultra · 128 GB | Q8_0 | Runs great | 46 GB | 11 |
Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.
Information
- Maker
- LG AI
- Released
- April 2026
- Parameters
- 33B
- Longest chat
- 256K tokens
- Reads pictures
- Yes
- Licence
- other
- Downloads this month
- 4,690
- Likes
- 55
- Sizes
- 6, 18 to 66 GB
- Kind
- Maker's release
Badge for the model card
Shows on EXAONE 4.5 33B'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.