Show HN: Vork – headless 3D and VFX engine for agents
I made an agent-native 3d/vfx engine. Agents are good at making interesting scenes in Blender/Unreal/Nuke/etc. But these tools are fundamentally limited compared to what coding agents are capable of. Many simplifying assumptions are baked into these tools in order to fit human authoring methods. In vork, agents write the whole scene as code, including their own shaders, materials, cameras and physics, and vork renders it on the GPU. The core API fits in about 2,500 tokens. So, for example, almos
Read original source
Source record
Recorded metrics
points · 5 votes
2026-11-10 00:45:16.464Z
cumulative since publication
Research analysis
Project record
Repository: voyager-so/vork. Source account: I made an agent-native 3d/vfx engine. Agents are good at making interesting scenes in Blender/Unreal/Nuke/etc. But these tools are fundamentally limited compared to what coding agents are capable of. Many simplifying assumptions are baked into these tools in order to fit human authoring methods. In vork, agents write the whole scene as code, including their own shaders, materials, cameras and physics, and vork renders it on the GPU. The core API fits in about 2,500 tokens. So, for example, almos
Observed adoption & comparisons
Source votes: 5 votes (Nov 10, 2026 UTC; cumulative since publication) Recent records to compare: “Show HN: Agent-email-guard – agent tried to send 400 receipts, stopped at 5” (Show HN, Nov 9, 2026); “Show HN: A Terminal UI for Hacker News, and CLI for Agents” (Show HN, Nov 9, 2026). Shared themes are navigation hints; they have not been verified as the same need or independent corroboration.
Adoption decision
Evaluate it against the tool or manual workflow it would replace. The stored record supports discovery; it does not yet establish reliability, lower operating cost or commercial adoption.
Unverified requirements
Not verified: license permissions, required hardware, installation steps, maintenance status and a representative task with its failure cases.
Next test
Choose one real task and a current alternative. Record the commit, environment, input, output, elapsed time and failure behavior; retain the result before recommending adoption.