Open-source dual-arm robot solves tangram puzzles with 90 demos, ~60% success
DanGrover · x · 2026-11-12
OpenWAM has been run on the low-cost open-source AlohaMini dual-arm robot: with fine-tuning on just 90 demonstration episodes, it achieved roughly a 60% success rate on a real-world tangram puzzle task, with full execution footage shown at 1x speed. Dan Grover is printing parts for an Aloha Mini 2 to try it himself.
More from Embodied
- Otto Pro: A Plug-and-Play Desktop Box That Runs AI Agents Locally, Keeping Your Data at Home — heyshrutimishra · 2026-11-12
- NVIDIA's Isaac Orbit Becomes Isaac Lab, Powering GR00T Robot Training Pipelines — animesh_garg · 2026-11-12
- Abbeel lab's SuperMimic replaces physics sim with differentiable world model, ~1000x faster sim-to-real — HirokatuKataoka · 2026-11-12
- Berkeley open-sources RoboSimStudio: AI agents program industrial robots without training data — berkeley_ai · 2026-11-12
- NXP ships pre-tested drone platform: initial flight in 30 minutes out of the box — shashib · 2026-11-12
- Robotics has three regional playbooks: China volume, Europe inspection, US pilots — ingliguori · 2026-11-12