MAPL
Python
★ 1
updated 3mo ago
Compact, non-Turing-complete DSL encoded as special tokens in LLMs, letting a model emit a full conditional tool-orchestration program (with branches, parallel calls, and silent exits) in a single output for a deterministic runtime to execute. cutting round-trips, tokens, and latency in multi-tool workflow
No plain-English explanation yet — one is being written right now. Check back in a minute.