workflow: inference hardcoded endpoint improvements

[hannahblair]

Description

• tweak remaining vision endpoints
• add custom param ui
• add (new?) image-text-to-video pipeline_tag for MiniMax H3

[abidlabs]

Could we add a couple of Workflow demos to the all_demos that are deployed when a Space is launched to make it easier to easily inspect the visual changes?

[abidlabs]

So for these other task types, should we automatically add the optional inputs? Like for vqa, we add an image, for dqa, we add a file input?

[hannahblair]

if im understanding your Q correctly, we already do via TASK_SCHEMAS in node-library.ts. though image to text was missing the optional text prompt so ive added that in

[hannahblair]

btw for dqa we use image not file because dqa routes through chat_completion, which only accepts types text and image_url :/

[hannahblair]

rather annoying UX

[hannahblair]

see here

[gradio-pr-bot]

🪼 branch checks and previews

•	Name	Status	URL
	Spaces	ready!	Spaces preview
	Website	ready!	Website preview
	Storybook	ready!	Storybook preview
🦄	Changes	failed!	Workflow log

---

Install Gradio from this PR

pip install https://huggingface.co/buckets/gradio/pypi-previews/resolve/438c55ca3915ab19f6b1511a3277eb9e93f9e25e/gradio-6.22.0-py3-none-any.whl

Install Gradio Python Client from this PR

pip install "gradio-client @ git+https://github.com/gradio-app/gradio@438c55ca3915ab19f6b1511a3277eb9e93f9e25e#subdirectory=client/python"

Import Gradio JS Client from this PR via CDN

import { Client } from "https://huggingface.co/buckets/gradio/npm-previews/resolve/438c55ca3915ab19f6b1511a3277eb9e93f9e25e/browser.js";