Skip to content

v1.1.0: Inline Schema Encoding

Choose a tag to compare

View all tags
@blackwell-systems blackwell-systems released this 13 Jun 06:12
· 14 commits to main since this release
v1.1.0
73b5f2b

v1.1.0: Inline Schema Encoding

The generic profile encoder now produces inline schema format for nested objects with 3+ scalar fields. This is a breaking change from v1.0.x output.

Breaking Changes

  • EncodeGeneric output format changed (inline schemas, no attachment indentation)
  • Decoder no longer accepts v1.0.x indented attachment syntax
  • Comma no longer quoted in pipe-delimited contexts
  • Strings starting with digits no longer quoted unless they match JSON number grammar

New Features

  • Inline object schema (^{fields}): nested objects encoded positionally
  • Shared array schemas: identical nested arrays omit field headers after first row
  • 25.5% fewer tokens than TOON across 15 real-world datasets
  • Codex CLI backend for eval harness (OpenAI models via ChatGPT subscription)
  • OpenAI Batch API support for eval

Integrity

  • 477M lossless round-trip iterations
  • 100% comprehension accuracy on every frontier model tested
  • 156 conformance fixtures passing
  • All error detection ported from v2 decoder (duplicate keys, orphan attachments, etc.)

Full spec: https://github.com/blackwell-systems/gcf/releases/tag/v3.0.0

Assets 2
Loading