AlpaSim
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 2
- NVIDIA releases Alpamayo 2 Edge autonomous-driving model
NVIDIA released Alpamayo 2 Edge on November 10, an end-to-end autonomous-driving VLA that combines causal reasoning, adaptive inference, and a diffusion action expert. The approximately 2.45B-parameter VLM backbone and 0.2B-parameter action expert run on DRIVE AGX Thor, with NVIDIA reporting a 48.9% closed-loop driving gain after reinforcement learning and a 160–190 ms Pixel-to-Trajectory latency in its reference configuration. The performance figures are vendor-reported and primarily reflect simulation-based evaluation.
雷峰网 AI · 🔥 0 - Qwen-Drive's 91.4 NAVSIM score is an oracle best-of-6 result
A Juejin AI technical audit finds that Qwen-Drive-1.0-RL's reported 91.4 PDMS score on NAVSIM v1.1 navtest comes from selecting the highest-scored trajectory among six candidates per scene using evaluation-time PDMS, while the default single trajectory scores 90.7. The analysis says this oracle result describes the potential of the candidate set rather than deployable selection ability, and it also examines evaluation denominators, time grids, data splits, and closed-loop evidence. The audit relied on official documentation, papers, and source code without running the model, simulator, scorer, or vehicle tests.
掘金 · 人工智能 · 🔥 0
Experience and discussion from the community
Share my AlpaSim experienceAsk about AlpaSim
Nobody has shared their experience with AlpaSim yet.