Favur Evals
favur.dev

The dataset

Every number on this site is generated from a validated, versioned canonical dataset — currently v1.0.0 (75 runs, rubric v15). An independent verification script recomputes sampled composites from the dataset's own component values on every publish.

Downloads

Static files on the site's CDN — cache-friendly and stable. Prior dataset versions stay retrievable at their versioned URLs. Checksums for every file are inmanifest.json.

Dataset downloads with file sizes
FileContentsSize
favur-evals-v1.0.0.jsonFull canonical dataset (immutable, versioned)1.5 MB
latest.jsonCurrent dataset version (stable URL)1.5 MB
run_subjects.csvFlattened CSV — one row per run × subject80 KB
run_agent_roles.csvFlattened CSV — one row per run × weighted agent role, with the model seat124 KB
schema.jsonJSON Schema (2020-12) for the full dataset10 KB
DATA_DICTIONARY.mdData dictionary — every field, unit, and semantics4 KB
manifest.jsonFile manifest with sha256 checksums
LICENSE.txtLicense (CC BY 4.0)568 B
CITATION.bibBibTeX citation405 B
runs/<sow>__<run>.jsonOne JSON document per run (75 files)1.5 MB

Reproduce it yourself

A standard-library-only Python script recomputes every composite from the public download's per-role component values and the published rubric block — no access to the scoring engine — and asserts the top run matches its published figure:reproduce_top_run.py. The same check runs in CI on every publish. Full formulas:methodology.

License & citation

The dataset is licensed underCC BY 4.0 — use it for anything, with attribution. Cite the dataset version and rubric version you used (currently dataset v1.0.0, rubric v15):

@misc{favurEvals,
  title        = {Favur Evals: a deterministic benchmark of LLMs inside a multi-agent software-development lifecycle},
  author       = {{Awesoft Solutions}},
  howpublished = {\url{https://evals.favur.dev}},
  note         = {Dataset: \url{https://evals.favur.dev/data}. Cite the dataset version and rubric version you used, e.g. dataset v1.0.0, rubric v15.},
  year         = {2026}
}

Pipeline health

The site republishes itself when new run data lands: validate → score → canonical dataset → fixture regression → public exports → deploy. Everything below derives from the data itself (never a hand-set date) and updates on every publish —health.json.

Publishing pipeline status
Runs published75
Latest ingested runsolarSystem/multi-xiaomi-mimo-v2.5-7.26.2026 (started 2026-07-25)
Runs skipped by the validation gate1
Dataset / rubric versionv1.0.0 / rubric v15

Changelog

Dataset versions, newest first
VersionDateRubricChanges
1.0.02026-08-01v15Initial canonical dataset: per-run composites, subject component breakdowns, per-role rubric cells with data-coverage fractions, telemetry aggregates, deliverable manifests, and entity catalogs (SoWs, vendors, models, agent roles). Scores identical to the pre-migration site (Sprint 0 baseline snapshot).

Every run on Favur Evals is scored by the same deterministic engine, on the same Statements of Work. The benchmark is self-funded — no vendor sponsorship, credits, or grants.