Is LTX-2.5 gated on Hugging Face? Agree, log in, then download
Yes: the LTX-2.5 repository is gated. Agree to share your contact details on the model page, run hf auth login, then hf download. It is free under $10M revenue.
Yes. The Hugging Face page for Lightricks/LTX-2.5 is gated: you must agree to share your contact information before you can access the files, and the card's download steps start with hf auth login. The weights are open, and the card says commercial use is free for companies under $10M in annual revenue. Sume has no LTX model id, so none of this gating applies to a Sume call, which needs only a Sume API key and a catalog id.
LTX details are from the LTX-2.5 model card and the LTX model page, read 2026-10-02. Sume's are from Video generation.
What do I have to do before I can download LTX-2.5?
The page shows a gate with the text that you need to agree to share your contact information to access the model, and a consent line about offers and updates. The card's download section then uses hf auth login followed by hf download Lightricks/LTX-2.5 with the file paths you want.
The card tags the license as ltx-2.x-community-license-agreement. Read it before you build on the weights; this post reports what the pages say and is not legal advice.
| Question | What the page says |
|---|---|
| Is it gated? | Yes; agree to share contact information to access the files |
| How do I authenticate? | hf auth login, then hf download |
| Under $10M annual revenue | Commercial and production use at no cost under the LTX-2.x Community License |
| Over $10M annual revenue | A paid commercial use agreement |
| Whose revenue counts? | The whole entity, including subsidiaries and affiliates under common control |
| Fine-tunes | Transfer of fine-tunes may require a paid license |
Does the gate apply to an API call?
The gate is on the weights repository. The LTX model page separately describes a managed API and an API playground, which this post does not cover. On Sume a video job needs Authorization: Bearer $SUME_API_KEY and a catalog id, and there is no LTX id in the OpenAPI model enum.
If you only need a clip, call a model Sume lists. If you need to run LTX-2.5 yourself, accept the gate once, log in on the machine that downloads, and keep the license terms with the project.
What is the quickest check that my login works?
Run hf auth whoami after logging in, then download one small file such as the video VAE before you pull the multi-gigabyte transformer. If the gate has not been accepted for your account, the download fails at that point instead of after a long transfer.
hf auth login
hf auth whoami
hf download Lightricks/LTX-2.5 \
vae/ltx-2.5-video-vae-bf16.safetensors \
--local-dir models/ltx-2.5Sources
Related posts
More in Models
- MAI-Image-2.6 output cap is 2,359,296 pixels; Sume sizes differ
MAI-Image-2.6 sets a 2,359,296-pixel ceiling and a 768-pixel minimum edge. Sume sets size per model with tiers, ratios and, for GPT models, custom pixels.
- MAI-Image-2.6 edits take 5 references; Sume takes 10 or 16
MAI-Image-2.6 in Foundry accepts up to five JPEG or PNG reference images per edit. On Sume, input_references tops out at 10, or 16 on GPT Image 2.5.
- MAI-Image-2.6 web_grounding flag: what Sume has instead
MAI-Image-2.6 can pull Bing results into an image when web_grounding is on. Sume has no such flag; here is how to pass current facts in the prompt instead.
- MiniMax H3 camera prompts: lens, movement, exposure wording
fal's H3 prompting guide says H3 reads film vocabulary: lens, rack focus, handheld, grain. Wording that works as a prompt and a Sume request that sends it.
Written by Sume