Hugging FaceFor making video clips that come with matching speech, background sound and sound effects.
- Parameters
- 22B
- Download
- 24 to 34 GB
- Commercial use
- With conditions
Tight fit on M4 · 16 GB
Outmake picks the Q4_K_S size: a 24 GB download that uses 8.3 GB of memory.
Made with LTX-2.3
See allThe rest are this month's best-liked on Civitai, shown from there; each belongs to the person who made it. Open one for its maker.
About
This version updates LTX-2, and Lightricks says it improves sound and picture quality and follows the request more closely. Lightricks released it with open weights and made it for running on your own computer.
Limitations
It is made for prompts written in English.
Sound without speech, like ambience or music, can come out lower in quality.
Results depend heavily on how the prompt is worded and may not match it fully.
Good to know
One part of the model makes the picture and a separate part makes the sound, side by side.
Sources: Lightricks/LTX-2.3 model card, LTX-2: Efficient Joint Audio-Visual Foundation Model. Every line comes from one of these.
Good at
Licence
LTX-2 Community License. Commercial use: with conditions. Free to use, for work too, for people and companies with under $10 million in yearly revenue. A bigger company needs a paid license from Lightricks. Its use policy bans, among other things, impersonating real people, false information and weapons work. The maker's page has the full terms.
LTX-2.3 runs well from a Mac with an M2 and 24 GB of memory, at Q4_K_S. 47 of 59 Mac configurations run it well.
Compare with
The nearest models in size, judged for the same Mac.
Sizes
| Size | Download | Memory at 8K | Open |
|---|---|---|---|
| Q4_K_SSmaller | 24 GB | 15 GB | Open |
| Q6_KBalanced | 29 GB | 19 GB | Open |
| Q8_0Best | 34 GB | 24 GB | Open |
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
LTX-2.3 is an open model. For ComfyUI, with the ComfyUI-GGUF nodes installed, put one of the GGUF files in ComfyUI/models/unet.
- ComfyUIWindows, Linux, Mac
Files
On every Mac
| Mac | Size | Fit | Memory |
|---|---|---|---|
| M6 · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M6 · 24 GB | Q4_K_S | Runs well | 15 GB |
| M6 · 32 GB | Q4_K_S | Runs great | 15 GB |
| M5 · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M5 · 24 GB | Q4_K_S | Runs well | 15 GB |
| M5 · 32 GB | Q4_K_S | Runs great | 15 GB |
| M5 Pro · 24 GB | Q4_K_S | Runs well | 15 GB |
| M5 Pro · 48 GB | Q6_K | Runs great | 19 GB |
| M5 Pro · 64 GB | Q8_0 | Runs great | 24 GB |
| M5 Max · 36 GB | Q4_K_S | Runs great | 15 GB |
| M5 Max · 48 GB | Q6_K | Runs great | 19 GB |
| M5 Max · 64 GB | Q8_0 | Runs great | 24 GB |
| M5 Max · 128 GB | Q8_0 | Runs great | 24 GB |
| M5 Ultra · 96 GB | Q8_0 | Runs great | 24 GB |
| M5 Ultra · 256 GB | Q8_0 | Runs great | 24 GB |
| M5 Ultra · 512 GB | Q8_0 | Runs great | 24 GB |
| M4 · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M4 · 24 GB | Q4_K_S | Runs well | 15 GB |
| M4 · 32 GB | Q4_K_S | Runs great | 15 GB |
| M4 Pro · 24 GB | Q4_K_S | Runs well | 15 GB |
| M4 Pro · 48 GB | Q6_K | Runs great | 19 GB |
| M4 Pro · 64 GB | Q8_0 | Runs great | 24 GB |
| M4 Max · 36 GB | Q4_K_S | Runs great | 15 GB |
| M4 Max · 48 GB | Q6_K | Runs great | 19 GB |
| M4 Max · 64 GB | Q8_0 | Runs great | 24 GB |
| M4 Max · 96 GB | Q8_0 | Runs great | 24 GB |
| M4 Max · 128 GB | Q8_0 | Runs great | 24 GB |
| M3 · 8 GB | Q4_K_S | Tight fit | 4.5 GB |
| M3 · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M3 · 24 GB | Q4_K_S | Runs well | 15 GB |
| M3 Pro · 18 GB | Q4_K_S | Tight fit | 8.3 GB |
| M3 Pro · 36 GB | Q4_K_S | Runs great | 15 GB |
| M3 Max · 36 GB | Q4_K_S | Runs great | 15 GB |
| M3 Max · 48 GB | Q6_K | Runs great | 19 GB |
| M3 Max · 64 GB | Q8_0 | Runs great | 24 GB |
| M3 Max · 96 GB | Q8_0 | Runs great | 24 GB |
| M3 Max · 128 GB | Q8_0 | Runs great | 24 GB |
| M3 Ultra · 96 GB | Q8_0 | Runs great | 24 GB |
| M3 Ultra · 256 GB | Q8_0 | Runs great | 24 GB |
| M3 Ultra · 512 GB | Q8_0 | Runs great | 24 GB |
| M2 · 8 GB | Q4_K_S | Tight fit | 4.5 GB |
| M2 · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M2 · 24 GB | Q4_K_S | Runs well | 15 GB |
| M2 Pro · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M2 Pro · 32 GB | Q4_K_S | Runs great | 15 GB |
| M2 Max · 32 GB | Q4_K_S | Runs great | 15 GB |
| M2 Max · 64 GB | Q8_0 | Runs great | 24 GB |
| M2 Max · 96 GB | Q8_0 | Runs great | 24 GB |
| M2 Ultra · 64 GB | Q8_0 | Runs great | 24 GB |
| M2 Ultra · 128 GB | Q8_0 | Runs great | 24 GB |
| M2 Ultra · 192 GB | Q8_0 | Runs great | 24 GB |
| M1 · 8 GB | Q4_K_S | Tight fit | 4.5 GB |
| M1 · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M1 Pro · 16 GB | Q4_K_S | Tight fit | 8.3 GB |
| M1 Pro · 32 GB | Q4_K_S | Runs great | 15 GB |
| M1 Max · 32 GB | Q4_K_S | Runs great | 15 GB |
| M1 Max · 64 GB | Q8_0 | Runs great | 24 GB |
| M1 Ultra · 64 GB | Q8_0 | Runs great | 24 GB |
| M1 Ultra · 128 GB | Q8_0 | Runs great | 24 GB |
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 LTX-2.3's Hugging Face card which Macs run it well, linked to this page. It updates every morning.
Sizes and files are read from unsloth/LTX-2.3-GGUF on Hugging Face and checked every morning. Catalog of November 4, 2026.