Ian Huang (@IanHuang3D)

2025-04-12 | โค๏ธ 28 | ๐Ÿ” 1


If youโ€™re wondering which multimodal LLMs you should be using to build 3D graphics agents ๐Ÿง‘โ€๐Ÿ’ป , check out our CVPR2025 Highlight work, BlenderGym โ€” not only does BlenderGym benchmark the top open and closed models, it also reveals a trick about how you should be allocating your inference compute for graphical editing tasks. With this trick, open source models can beat close-source models on 3D graphics editing. Curious? ๐Ÿง ๐Ÿ‘‰ https://blendergym.github.io/

๐Ÿ”— ์›๋ณธ ๋งํฌ


Auto-generated - needs manual review

์ธ์šฉ ํŠธ์œ—

Yunqi (Richard) Gu (@richard_yunqigu)

Which multimodal LLM should you be using to edit graphics in Blender?

Today, weโ€™re releasing our CVPR2025 Highlight๐ŸŒŸ work, BlenderGym ๐Ÿ‹๏ธโ€โ™€๏ธ, the first agentic 3D graphics editing benchmark that will tell you exactly how multimodal LLMs compare in their Blender-editing skills.

Whatโ€™d we find? ๐Ÿงต๐Ÿ‘‡

์›๋ณธ ํŠธ์œ—

๐ŸŽฌ ์˜์ƒ

Tags

domain-vision-3d domain-rendering domain-ai-ml domain-vlm domain-visionos