Show HN: Blender for AI Agents (mixar.app)

6 points by namanbhargava18 ↗ HN
Creating 3D is hard. LLMs seem to be getting better at tool use and spatial understanding. While MCPs have proved to be a good way to use these tools- the current methods have these challenges:

- Access to scene graph and core C modules of Blender

- Lack of parallelism, only way is to run blender headless

- Lack of deterministic and fast verification layer

- Inference stack- only way to use inference is to hook another MCP

We're building Mixar, think Cursor for 3D. One access point to all generative inference, an agent to build scenes/blockouts, do boring stuff like UVs and export standard formats(.glb/gltf/obj/gbx/usd etc.)

Try it: https://mixar.app - first week is free. Would love to get your feedback.

7 comments

[ 5.1 ms ] story [ 27.0 ms ] thread
I have tried a lot of AI 3D stuff but nothing is usable in actual games. Sometimes it makes my character rigs so weird that I am unable to animate it.
I’m so bullish on the idea that LLMs will solve spatial understanding themselves. World model makers can succ ma d.
I’m a bit skeptical about LLMs actually being able to perceive 3D context properly. I think a well-designed structured scene graph could be promising.
Does the agent have any guardrails around scene complexity before it makes changes?