Hugging Face Free Video Generation

Questions
1

Hi!
I tried using Hugging face api for video generation but it is actually giving me errors, It says to use its new “router” link, but when I use that it says there are not enough credits or something. It was a bit unrealiable. Do you guys have a workflow of hugging face that I can try?

Describe the problem/error/question

What is the error message (if any)?

Please share your workflow

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:
2

Hey! The errors you’re seeing are because Hugging Face deprecated their old Serverless Inference API (api-inference.huggingface.co) in late 2025. That’s why it’s returning an error and telling you to use its “router” instead.

For video generation specifically in n8n, I’d honestly recommend skipping HF’s inference API and going with a dedicated video API instead. A few options that work well with n8n’s HTTP Request node include: Runway ML API, Replicate and fal.ai.

The pattern in n8n is basically: HTTP Request node (POST to start generation) → Wait node (15-60 sec depending on model) → HTTP Request node (GET to check status/retrieve result).

3

Hy @azizaavon

Look if it help

4

Hi @azizaavon!

You can try this sample workflow i have found:

Hope this helps

5

@azizaavon For testing Hugging Face, start with image generation (stable-diffusion) first to verify your setup works, then move to video.

6

Hi @Anshul_Namdev
Tried using that model but it throws an error of “No Inference Provider available for model mistralai/Mistral-7B-Instruct-v0.1.”
Which means that the Hugging Face serverless API endpoint is overloaded, the model is too large to load automatically (e.g., >10GB), or the model is not currently hosted by the free API tier

7

Oh.. I was hoping for free alternatives tbh
All of the listed are pay to use with no trial either…
But Thank you

8

@azizaavon, I’d recommend trying Hugging Face Spaces directly instead of the API. Models like Pyramid Flow SD3 run on free ZeroGPU. You can use n8n’s HTTP Request node to interact with the Space endpoint.Replicate also gives $10 free credits on signup with no credit card needed, which should cover several test videos.

9

If you are looking for FREE models do not consider hugging face, as only very low parameter models are available for free, use a service like OpenRouter || Groq, and you are here generating videos so please do not rely on free tiers and if you really want to test things out, consider:

Hope this helps.