OPEN MODELS / AI music and audio
Run LTX-2.3-22b-LoRA-Foley-V2A on a cloud GPU
LTX-2.3-22b-LoRA-Foley-V2A is a AI music and audio model by Lightricks. See how much GPU memory it needs, which GPUs it fits on today and what it costs to run it yourself, billed per second.
GPUs it fits on today
Our estimate is about 54 GB for inference with BF16 weights, including working memory. Real use depends on resolution, batch size and context length.
| GPU | Memory | Per hour | 10 hours | |
|---|---|---|---|---|
| CMP 170HX | 64 GB | $0.62 | $6.20 | Rent |
| GB10 | 119 GB | $0.71 | $7.10 | Rent |
| A100 SXM4 | 80 GB | $0.72 | $7.20 | Rent |
| RTX PRO 6000 Max-Q | 96 GB | $1.72 | $17.20 | Rent |
| RTX PRO 6000 WS | 96 GB | $2.36 | $23.60 | Rent |
Live starting prices from our catalog. You pay per second while the machine runs.
How to run LTX-2.3-22b-LoRA-Foley-V2A
- 01
Rent a GPU
Pick one of the GPUs above. Billing is per second; stop the machine and you stop paying.
- 02
Open a terminal or Jupyter
Connect over SSH or open Jupyter in your browser. PyTorch and CUDA are ready to use.
- 03
Download and run the model
Run the commands below. The weights download straight from Hugging Face to your machine.
This model is gated: accept its license on Hugging Face and run huggingface-cli login before downloading.
pip install -U transformers accelerate
python -c "from transformers import pipeline; p = pipeline('text-to-audio', model='Lightricks/LTX-2.3-22b-LoRA-Foley-V2A', device=0); print(p)"Why run it yourself?
Make your own songs and sound without a subscription or per-track fee. Your GPU, your files.
Questions
- How much GPU memory does LTX-2.3-22b-LoRA-Foley-V2A need?
- By our estimate about 54 GB for inference (22B parameters). Higher resolution, longer context or bigger batches need more.
- How much does it cost to run LTX-2.3-22b-LoRA-Foley-V2A?
- From $0.62 per hour on CMP 170HX, billed per second. Ten hours cost about $6.20.
- Do I need a subscription?
- No. You pay only for the time your machine runs, from your USD balance. Stop it whenever you are done.
Try it on your own GPU in minutes
Pick a GPU, open Jupyter or SSH and run any open model. You pay per second and can stop anytime; no subscription, no setup fee.
- Per-second billing in USD
- PyTorch and CUDA ready in minutes
- Rent from ChatGPT or Claude too