Hedra MCP + CLI are live: the first step toward our platform for building, improving, and deploying…
@hedra_labs54 тыс. подписчиков
96 348 просмотров
36 комментариев
Комментарии из оригинального поста в X (Twitter)

@hedra_labs
Start building today:
21 отметка «Нравится»

@mint0themax
congrats!!
2 отметки «Нравится»

@henloitsjoyce
alright dropping my demo here first - to the people trying it out - show us whatchu got!!
2 отметки «Нравится»

@xPolarrr
lfg this is sick
2 отметки «Нравится»

@hedra_labs
it's about to be a movieeee

@haroonisdreamin
congrats!!
2 отметки «Нравится»

@Everett_Butler
One small step for Hedra. One giant leap for develoeprs.
2 отметки «Нравится»

@andrewchen
can't wait to try this @mjlbach - congrats to the Hedra team
2 отметки «Нравится»

@hedra_labs
@mjlbach show us your creation 👀

@Kantorcodes
@hedra_labs An MCP that puts image, video and world-model generation inside agent loops is a clean action surface for HOL Guard. Hedra maps to contributions/mcp-servers/mcp.hedra.json:
1 отметка «Нравится»

@zsakib_
congrats, good to see you infront of the camera! 😄
1 отметка «Нравится»

@jigglypuffmeow4
🫡
1 отметка «Нравится»

@flhnara
This is a huge unlock for agentic visual workflows. Bringing Hedra’s image, video, and world models directly into the tools developers already use feels like the right direction. Excited to try the MCP + CLI.
1 отметка «Нравится»

@skeptrune
you, uh, need some inference?
1 отметка «Нравится»

@TheSITherapist
Hedra now serves Claude Code and ChatGPT via MCP. I like how easy it is to build vision models this way. You should try connecting your agent today.
1 отметка «Нравится»

@baskojpeg
Can I pass character and location reference images through MCP when generating video? That’s the first thing I’d test for a short film.
1 отметка «Нравится»

@tsaniauly1
I'm so excited, it's going to be much easier now! 🔥
1 отметка «Нравится»

@DjajaYerry75
The new Hedra ability to connect to many different LLM's will add more flexibility to many users.
1 отметка «Нравится»

@PrettyFavy17
The workflow gets a lot more powerful when creation is built in.

@reezanahamed
This is the direction I think explainer video needs, code calling the model instead of typing prompts each time. I'm building something similar in the open, an engine where any LLM writes the video through code. Curious how you're handling versioning when an agent iterates on the same clip a bunch of times.
















