
ORA
Open models operate robots. Zero-shot.
ORA is an open harness for robot use. The model looks, reasons and writes programs; a supervisor admits every motion; every run is sealed and can be replayed.
One recorded run for each task ORA v0.1 solves: 27 of RoboDojo's 40 tasks and 90 of RoboLab's 100, each marked as it succeeds. DeepSeek-V4.1-Flash, code mode, zero-shot; sped up.
A free-play session in code mode, replayed from its records: the operator asks for a tower and changes the order halfway. Left, the programs the agent writes and what comes back; right, the cameras. DeepSeek-V4.1-Flash, RoboDojo toy table, unscored.
v0.1What the agent does now
Four changes since October's preview. Each opens its part of the post.
- Write code.The model writes Python against the same robot tools; one turn carries 3.6 robot actions instead of 1.1.
- Move smoothly.Every move is planned around what the depth cameras see; moves stopped by a collision fall from 14.1% to 7.0% on RoboDojo.
- Work across open models.In code mode on 39 RoboDojo tasks, six open models solve from 48.7% (DeepSeek) down to 5.1% (Nemotron).
- Cold-start post-training.On RoboLab-120, Qwen3.8-27B solves 5.4% alone, 52.5% inside ORA, and 61.9% with PEAR's training on top.
ResultsZero-shot, beside published systems
ORA in code mode with DeepSeek-V4.1-Flash, an open-weight model, on the cases of RPent's leaderboard.
RoboTwin C2R
LIBERO-Pro
ORA, zero-shotRPent: LLM + VLA + task memoryVLA trained on demonstrations
Rows other than ORA are RPent's published leaderboard (snapshot of 2026-09-22). ORA: 235 of RPent's 250 RoboTwin cases; LIBERO-Pro Task + Swap, initial states 0–4. RoboLab, RoboDojo and the same model in other systems are in the post.
MethodHow ORA works
The model asks for an action; the harness carries it out on the robot and returns what happened; the model decides again. A turn carries one tool call, or one program.
The agent loop, around a robot
Tool calls. Each turn the model sends one tool call; the harness runs it and one result comes back before the model decides again.
Code mode. Each turn the model writes one program, which calls the same tools from a persistent workspace, up to 12 robot motions admitted one at a time, before one result comes back.
Run it
One harness for RoboTwin, LIBERO-Pro, RoboDojo, RoboLab and a real arm, with any OpenAI-compatible or Anthropic model.
curl -fsSL https://raw.githubusercontent.com/fast-rua/ORA/preview/v0.1/scripts/install.sh | sh -s -- --ref preview/v0.1
ora providers key DEEPSEEK_LOW_API_KEY # any OpenAI-compatible or Anthropic endpoint ora play --env synthetic # no GPU: give a robot instructions
ora setup --env robodojo # simulator, assets, sandbox ora run --env robodojo --task general_pickup --confirm-launch