SAVRN
Search Contact SAVRN

Open-weight model · Text to video

LTX-2.5-Diffusers

by LTX.io Lightricks/LTX-2.5-Diffusers

LTX-2.5 is an open world model with open weights, built for local execution and fine-tuning.

Parameters19B
Context—
Weights174.4 GB
Licenseother
AccessAccess requested at publisher
Monthly Downloads105k

Runs On

What it takes to serve LTX-2.5-Diffusers (19B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 38.0 GB 45.6 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 19.0 GB 22.8 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 9.5 GB 11.4 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

Model Card

LTX-2.5 is an open world model with open weights, built for local execution and fine-tuning. Its established use is generating synchronized, high-fidelity video and audio from text, image, and video inputs; applicability to emerging domains such as robotics and physical AI is developing. Full control and customization — self-host on your own infrastructure. No per-generation billing, no per-seat lock-in, no forced API dependency. Revenue is measured across the whole entity, including subsidiaries and affiliates under common control. The full, binding terms live in LICENSE. Encoding always uses vae/, and LTX2Pipeline decodes with vae/ too. The diffusion decoder is a diffusion model in its…

Excerpt from the card by LTX.io, licensed other.

Identity and Version

Repository
Lightricks/LTX-2.5-Diffusers
Publisher
LTX.io
Task
Text to video
Modality
Video
Library
diffusers
Parameters
19B parameters
Languages
en, de, es, fr, ja, ko, zh, it
Revision
426936f8b22dc28e4def61e515478b0b7e4a53cc
First published
2026-07-26
Last updated
2026-08-24

Files and Weights

66 files, 174.4 GB in total. The weights are 35 files totalling 174.4 GB in safetensors.

Weights35 files · 174.4 GB
Configuration23 files · 1.2 MB
Tokenizer4 files · 64.3 MB
Documentation1 file · 29.9 KB
Other2 files · 37.3 KB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
audio_vae/diffusion_pytorch_model.safetensorsWeights106.5 MB —
connectors/diffusion_pytorch_model-00001-of-00002.safetensorsWeights5.0 GB —
connectors/diffusion_pytorch_model-00002-of-00002.safetensorsWeights1.3 GB —
connectors/diffusion_pytorch_model.safetensorsWeights6.3 GB —
diffusion_decoder/diffusion_pytorch_model.safetensorsWeights834.3 MB —
duration_head/diffusion_pytorch_model.safetensorsWeights3.8 MB —
latent_upsampler/diffusion_pytorch_model.safetensorsWeights995.7 MB —
ltx-2.5-22b-distilled-lora-450-bf16.safetensorsWeights9.7 GB —
prompt_enhancer/model-00001-of-00003.safetensorsWeights1.4 GB —
prompt_enhancer/model-00002-of-00003.safetensorsWeights5.0 GB —
prompt_enhancer/model-00003-of-00003.safetensorsWeights3.8 GB —
temporal_latent_upsampler/diffusion_pytorch_model.safetensorsWeights261.9 MB —
text_encoder/model-00001-of-00005.safetensorsWeights5.0 GB —
text_encoder/model-00002-of-00005.safetensorsWeights5.0 GB —
text_encoder/model-00003-of-00005.safetensorsWeights4.9 GB —
text_encoder/model-00004-of-00005.safetensorsWeights5.0 GB —
text_encoder/model-00005-of-00005.safetensorsWeights4.1 GB —
transformer/diffusion_pytorch_model-00001-of-00004.safetensorsWeights9.9 GB —
transformer/diffusion_pytorch_model-00001-of-00008.safetensorsWeights5.0 GB —
transformer/diffusion_pytorch_model-00002-of-00004.safetensorsWeights9.9 GB —
transformer/diffusion_pytorch_model-00002-of-00008.safetensorsWeights4.9 GB —
transformer/diffusion_pytorch_model-00003-of-00004.safetensorsWeights10.0 GB —
transformer/diffusion_pytorch_model-00003-of-00008.safetensorsWeights5.0 GB —
transformer/diffusion_pytorch_model-00004-of-00004.safetensorsWeights8.1 GB —
transformer/diffusion_pytorch_model-00004-of-00008.safetensorsWeights4.9 GB —
transformer/diffusion_pytorch_model-00005-of-00008.safetensorsWeights5.0 GB —
transformer/diffusion_pytorch_model-00006-of-00008.safetensorsWeights5.0 GB —
transformer/diffusion_pytorch_model-00007-of-00008.safetensorsWeights5.0 GB —
transformer/diffusion_pytorch_model-00008-of-00008.safetensorsWeights3.2 GB —
transformer_full/diffusion_pytorch_model-00001-of-00004.safetensorsWeights9.9 GB —
transformer_full/diffusion_pytorch_model-00002-of-00004.safetensorsWeights9.9 GB —
transformer_full/diffusion_pytorch_model-00003-of-00004.safetensorsWeights10.0 GB —
transformer_full/diffusion_pytorch_model-00004-of-00004.safetensorsWeights8.1 GB —
vae/diffusion_pytorch_model.safetensorsWeights1.5 GB —
vocoder/diffusion_pytorch_model.safetensorsWeights258.3 MB —
audio_vae/config.jsonConfiguration505 B —
connectors/config.jsonConfiguration817 B —
connectors/diffusion_pytorch_model.safetensors.index.jsonConfiguration30.6 KB —
diffusion_decoder/config.jsonConfiguration1.1 KB —
duration_head/config.jsonConfiguration251 B —
latent_upsampler/config.jsonConfiguration301 B —
model_index.jsonConfiguration971 B —
modular_model_index.jsonConfiguration3.9 KB —
processor/processor_config.jsonConfiguration1.7 KB —
prompt_enhancer/config.jsonConfiguration5.3 KB —
prompt_enhancer/generation_config.jsonConfiguration204 B —
prompt_enhancer/model.safetensors.index.jsonConfiguration200.0 KB —
scheduler/scheduler_config.jsonConfiguration489 B —
temporal_latent_upsampler/config.jsonConfiguration299 B —
text_encoder/config.jsonConfiguration4.6 KB —
text_encoder/generation_config.jsonConfiguration255 B —
text_encoder/model.safetensors.index.jsonConfiguration66.3 KB —
transformer/config.jsonConfiguration1.4 KB —
transformer/diffusion_pytorch_model.safetensors.index.jsonConfiguration446.5 KB —
transformer_full/config.jsonConfiguration1.4 KB —
transformer_full/diffusion_pytorch_model.safetensors.index.jsonConfiguration446.5 KB —
vae/config.jsonConfiguration1.5 KB —
vocoder/config.jsonConfiguration1.5 KB —
README.mdDocumentation29.9 KB —
processor/chat_template.jinjaOther18.6 KB —
tokenizer/chat_template.jinjaOther18.7 KB —
.gitattributesRepository1.6 KB —
processor/tokenizer.jsonTokenizer32.2 MB —
processor/tokenizer_config.jsonTokenizer3.7 KB —
tokenizer/tokenizer.jsonTokenizer32.2 MB —
tokenizer/tokenizer_config.jsonTokenizer3.7 KB —

License and Download

License
other
Access
Access requested at publisher
Download size
174.4 GB
Request access from LTX.io

LTX.io grants access through its official repository on Hugging Face. Read the license.

Built From

Memory Requirements

PrecisionWeights in memory
As published174.4 GB
16-bit38.0 GB
8-bit19.0 GB
4-bit9.5 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About LTX-2.5-Diffusers

How much GPU memory does LTX-2.5-Diffusers need?

About 45.6 GB at 16-bit and 11.4 GB at 4-bit: the weights (19B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run LTX-2.5-Diffusers on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

What license is LTX-2.5-Diffusers released under?

other, as its publisher declares it. Read the license text before commercial use.

Similar Models

We are excited to introduce Wan2.2, a major upgrade to our foundational video models. With Wan2.2, we have focused on incorporating the following innovations: This repository contains our T2V-A14B model, which supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions. Your browser does not support the video tag. If your research or project builds upon Wan2.1 or Wan2.2, we welcome you to share it with us so we can highlight it for the broader community. - Wan2.2…

Open weights apache-2.0 14.3B parameters diffusers
View model

Model · Text to video

Wan2.2-T2V-A14B-Diffusers

Wan-AI

We are excited to introduce Wan2.2, a major upgrade to our foundational video models. With Wan2.2, we have focused on incorporating the following innovations: This repository contains our T2V-A14B model, which supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions. Your browser does not support the video tag. If your research or project builds upon Wan2.1 or Wan2.2, we welcome you to share it with us so we can highlight it for the broader community. - Wan2.2…

Open weights apache-2.0 14.3B parameters diffusers
View model

Model · Text to video

Wan2.1-T2V-14B-Diffusers

Wan-AI

In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features: This repository features our T2V-14B model, which establishes a new SOTA performance benchmark among both open-source and closed-source models. It demonstrates exceptional capabilities in generating high-quality visuals with significant motion dynamics. It is also the only video model capable of producing both Chinese and English text and supports video generation at both 480P and 720P resolutions. Your browser does not support the video tag. - Wan2.1 Text-to-Video - [x] Multi-GPU Inference code of the 14B and 1.3B…

Open weights apache-2.0 14.3B parameters diffusers
View model

Model · Text to video

Wan2.1-T2V-14B

Wan-AI

In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features: This repository features our T2V-14B model, which establishes a new SOTA performance benchmark among both open-source and closed-source models. It demonstrates exceptional capabilities in generating high-quality visuals with significant motion dynamics. It is also the only video model capable of producing both Chinese and English text and supports video generation at both 480P and 720P resolutions. Your browser does not support the video tag. - Wan2.1 Text-to-Video - [x] Multi-GPU Inference code of the 14B and 1.3B…

Open weights apache-2.0 14.3B parameters diffusers
View model

Model · Text to video

Wan2.2-TI2V-5B-Diffusers

Wan-AI

We are excited to introduce Wan2.2, a major upgrade to our foundational video models. With Wan2.2, we have focused on incorporating the following innovations: This repository contains our TI2V-5B model, built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research. Your browser does not support the video tag. If your research or project builds upon Wan2.1 or Wan2.2, we welcome you to share it…

Open weights apache-2.0 5B parameters diffusers
View model

You can try our models here！ We're excited to introduce the FastWan2.2 series—a new line of models finetuned with our novel Sparse-distill strategy. This approach jointly integrates DMD and VSA in a single training process, combining the benefits of both distillation to shorten diffusion steps and sparse attention to reduce attention computations, enabling even faster video generation. FastWan2.2-TI2V-5B-Full-Diffusers is built upon Wan-AI/Wan2.2-TI2V-5B-Diffusers. It supports efficient 3-step inference and produces high-quality videos at 121×704×1280 resolution. For training, we used simulated forward for the generator model, making the process data-free. The current…

Open weights apache-2.0 5B parameters diffusers
View model