Introducing Artifacts for Orion 🧵
https://www.vlm.run/blog/introducing-orion-artifacts
Multi-modal outputs as first-class objects. No URL juggling. No storage glue.
Generate images, videos, masks, and 3D reconstructions – and get typed references you can store, reuse, and compose.
Real multi-modal pipelines chain steps: detect → crop → redact → annotate → generate.
The hardest part isn't the model anymore. It's the glue code that moves media between steps.
Temporary URLs, signed URLs, external storage... it adds latency and breaks down fast.
What you can build with Artifacts:
• Generate multiple image views for virtual try-on in a single call
• Edit images or videos and return multi-modal outputs
• Generate 3D assets from images
• Any multi-step pipeline where intermediate media needs to be reusable
Available now for all Orion API users.
📖 Blog: https://vlm.run/blog/introducing-orion-artifacts
👨💻Cookbook: https://github.com/vlm-run/vlmrun-cookbook
📚 Docs: https://docs.vlm.run/agents/artifacts
VLM Run