Companion resources for the Gemini vision labeling loop: SAM3, Falcon docs
mrdbourke · x · 2026-10-23
mrdbourke shared a resource list accompanying his Gemini 4 Argon vision demo:
- Gemini vision docs: how to get boxes/masks on images
- Gemini video docs: using agentic video for video understanding/editing (he uses it to sort his raw footage)
- SAM3: high-quality masks (Gemini for boxes → SAM3 for masks)
- Falcon Perception / Falcon Perception-HD: lightweight local models, noticed to be better at specific text-based prompts for at-scale local runs
Together with the main post it forms a complete agent vision-labeling workflow reference.
More from coding & agent
- Dev demos AI models handling 3D objects and animations surprisingly well — chongdashu · 2026-10-23
- Jason Kneen burns 104.56B tokens and $24.7K in 304 days of AI coding on Codex and Claude Code — jasonkneen · 2026-10-23
- Typosquatted Blender MCP Repo Hit 958 Stars in Days and Silently Ran Malware — Promptmethus · 2026-10-23
- Benchmarking 56 fresh dependency bugs: MCP routing lifts Claude Code accuracy 68% to 91% — Available_Side_7134 · 2026-10-23
- Swapping only the eval harness moves results 30%: designing agent evals like a scientist — AI Engineer · 2026-10-23
- Med student builds a whole-library RAG with Cognee; single-book ingestion still freezes his machine — shrike293103 · 2026-10-23