← Leaderboard
LLaRA
VLAStony Brook University · Jun 28, 2024
Framework from Stony Brook University and UW–Madison that turns robot trajectories into visuo-textual conversation data and instruction-tunes LLaVA-1.5-7B into a policy. It is evaluated on VIMA-Bench and on real-robot tasks, not on a tracked benchmark. The Apache-2.0 license covers the code; the checkpoints inherit LLaVA’s terms.
- Type
- VLA
- Parameters
- 7B
- Backbone
- LLaVA-1.5-7B
- Weights
- Open
- License
- Apache-2.0
- Commercial use
- Unclear
- Released
- Jun 28, 2024
- Robots
- VIMA-Bench (sim), xArm
- SO-101
- –
- LeRobot
- –
- Inputs
- RGB · L
- VRAM
- 24GB (minimum)
Benchmark results
0 of 13 benchmarksNo benchmark results recorded yet.
Not reported on 13 benchmarks
Wrong or missing number? Report an error