An honest walk through what FAST compresses, why the compression works, and why only half of the paper survived.
FAST is best understood as a research arc, not a single result. In January 2025, Physical Intelligence released it as an autoregressive alternative to flow matching — a tokenizer plus an AR runtime head that would let a 3B VLM emit dexterous actions directly. By May 2025, PI had retired the runtime architecture: inference was roughly 2× slower than flow matching, and deployment pathologies stacked up. What stayed load-bearing is the tokenizer itself, now the standard pretraining auxiliary target inside Knowledge Insulation and π₀.5. This workshop walks that arc honestly.
Sources: FAST §1 (arXiv 2501.09747); KI paper §intro (arXiv 2505.23705).