LBM Eval
Foundry-Qwen3VLA-2.1B 72.4% Fading
Bimanual robot policies run 49 manipulation tasks in the Drake simulator from random starting conditions. Scored by success rate per task.
About 125 GitHub stars. Version 1.1.0 (fall 2025) is newer than the snapshot the LBM 1.0 paper used, and the README says it may not reproduce the paper. TRI's VLA Foundry cards score 16 seen and 3 unseen tasks.
Made by Toyota Research Institute Released May 2025 Simulated robots All tasks public Last in a launch post Apr 2026 , dropped before it saturated
Last model added Apr 2026
- Level
- Everyday Most adults, without training
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Agent task Many steps in a terminal, browser, desktop, app, game or robot.
2
Models scored in lab cards
2
Launch posts list it, from 1 lab
Foundry-Qwen3VLA-2.1B leads with 72.4%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
- 1 Foundry-Qwen3VLA-2.1B 72.4%
- 2 Foundry-VLA-1.7B 34.1%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Foundry-Qwen3VLA-2.1B | 72.4% | Apr 2026 | Model card | |
| 2 | Foundry-VLA-1.7B | 34.1% | Apr 2026 | Model card |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Reported by the lab in its model card
Labs headline it less often now, last in Apr 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 21 Apr 2026 | Foundry-VLA-1.7B | 34.1 | OSS simulator, success rate on 16 seen tasks, 200 rollouts per task. Unseen (3 tasks) 7.5% (Evaluation Results table) |
| 21 Apr 2026 | Foundry-Qwen3VLA-2.1B | 72.4 | OSS simulator, success rate on 16 seen tasks, 200 rollouts per task. Unseen (3 tasks) 13.5% (Evaluation Results table) |
More in Robotics
- ERQA In use 400 multiple-choice visual questions from robot and egocentric footage on spatial, trajectory, action and state reasoning. Accuracy.
Gemini Robotics ER 2 78.5% - RoboLab In use Zero-shot DROID Franka policies in Isaac Lab on 120 tabletop tasks at three instruction levels, 10 episodes each. Success rate.
GR00T N1.7 8.58% - Human proximity (Google DeepMind robotics) In use Models judge from robot camera views whether a human is within 1 meter, a spatial awareness check for safe robot behavior. Scored by accuracy. Gemini Robotics ER 2 93%
- Moment finding (Google DeepMind) In use Models find the moment in a robot video when an event happens. Scored by accuracy and mean absolute time distance in seconds. Gemini Robotics ER 2 91.3%
- Progress classification (Google DeepMind) In use Models assign each frame of a robot video to one of five task-progress bands from 0 to 100 percent. Scored by accuracy. Gemini Robotics ER 2 57.4%
- LIBERO Fading Simulated Franka tabletop tasks in four 10-task suites. Policies fine-tuned on 50 demos per task. Success rate over ~50 trials.
MolmoAct 2 97.2%