test: add end-to-end token optimization benchmark

Browse source

Benchmark validates all optimization modules with realistic GSD content:
- Structured data: 20% decisions savings, 7% requirements savings
- Prompt compression: 5-17% across light/moderate/aggressive levels
- Semantic chunking: 73% content reduction via TF-IDF selection
- Summary distillation: 73% savings preserving structured fields
- Combined pipeline: 43% total savings on realistic dispatch prompt
- Cache efficiency: 94% cacheable prefix, 85% estimated Anthropic savings
- Provider-aware: 14% budget accuracy improvement for Anthropic vs OpenAI

...

This commit is contained in:

Jeremy McSpadden 2026-03-17 22:10:58 -05:00

parent d65da6c927

commit 4e7b3d486f

1 changed files with 1272 additions and 0 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Show stats Download patch file Download diff file Expand all files Collapse all files

1272

src/resources/extensions/gsd/tests/token-optimization-benchmark.test.ts Normal file

View file

File diff suppressed because it is too large Load diff