Jon Saad-FalconandClaude Opus 4.6 f6c02c13d8 feat(evals): add --compact and --trace-detail CLI flags ...

Wire new display flags through the run command to print_full_results.
--compact renders a single dense table; --trace-detail enables full
per-step trace listing.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-02-28 17:18:11 +00:00

..

__init__.py

Add evaluation framework and center README logo SVGs

2026-02-22 23:48:43 +00:00

conftest.py

Add eval framework efficiency metrics: MFU, MBU, IPW, IPJ

2026-02-25 21:17:53 +00:00

test_backends.py

Add evaluation framework and center README logo SVGs

2026-02-22 23:48:43 +00:00

test_cli_flags.py

feat(evals): add --compact and --trace-detail CLI flags

2026-02-28 17:18:11 +00:00

test_config.py

Add eval framework efficiency metrics: MFU, MBU, IPW, IPJ

2026-02-25 21:17:53 +00:00

test_display.py

feat(evals): grouped display tables (accuracy, latency, energy, trace)

2026-02-28 17:13:04 +00:00

test_frames.py

Add evaluation framework and center README logo SVGs

2026-02-22 23:48:43 +00:00

test_gaia.py

Add evaluation framework and center README logo SVGs

2026-02-22 23:48:43 +00:00

test_runner.py

Wire trace data and token totals into eval RunSummary (Task 6)

2026-02-28 17:03:15 +00:00

test_supergpqa.py

Add evaluation framework and center README logo SVGs

2026-02-22 23:48:43 +00:00

test_types.py

Add trace and power fields to EvalResult and RunSummary (Task 5)

2026-02-28 07:20:21 +00:00

test_wildchat.py

Add evaluation framework and center README logo SVGs

2026-02-22 23:48:43 +00:00