NF-VLA
VLA from the University of Pennsylvania and the Australian National University that generates action chunks with an autoregressive normalizing flow conditioned on VLM features, as a likelihood-based alternative to diffusion and flow-matching heads.
University of Pennsylvania & ANU · Nov 15, 2026 · VLAReport an error
98.3%LIBERO · #35 of 83
74.2%LIBERO+ · #27 of 37
2Benchmarks reported
Nov 15, 2026Released
Benchmark results2
LIBERO-PlusSimulation
- Paper / model card ↗Nov 2026
#27 of 3774.2%, self-reported by the authors, not a leaderboard entry
- Paper / model card ↗Nov 2026· n = 2000
#35 of 8398.3%, self-reported by the authors, not a leaderboard entry
Not reported on 13 benchmarks
Robustness drop: 98.3% on LIBERO → 74.2% on LIBERO-Plus. The perturbed variants show how much of the LIBERO score is memorization.
Running the model
| Hardware | Precision | Config | Latency / chunk | Steps | VRAM | Chunk | Control | Source |
|---|---|---|---|---|---|---|---|---|
| RTX 4080 SUPER | – | FP32 with BF16 autocast | 74.1 ms | – | – | 10 | – | Nov 2026 ↗ |
Wrong or missing number? Report an error