Commit History

  1. Clarify reward-transfer inference scope

    Verified signature
    42b53c3
  2. Add verified reward-transfer results

    Verified signature
    ae141be

  1. Align Space card with revised evidence⁠

    Verified signature
    aae8e76

  1. Publish the arXiv-linked v1.3 leaderboard⁠

    Verified signature
    f0c6805

  1. Sync FlavourBench v1.2.0 source

    Verified signature
    6aec8cd
  2. Remove injected host chrome from the benchmark canvas

    Verified signature
    cce45fc
  3. Add task-family leaderboard views and research navigation

    Verified signature
    81e6eb1
  4. Remove decorative framing and table border leaks

    Verified signature
    3c3ecdd
  5. Align hosted Space chrome with launch design

    Verified signature
    2c640ad
  6. Polish benchmark discovery and statistical insights

    Verified signature
    2de538e
  7. Match Space to launch design

    Verified signature
    ad13bd6
  8. Sync launch asset to repository source

    Verified signature
    1d92e52
  9. Redesign Space and surface turnkey model evaluation⁠

    Verified signature
    418bf00

  1. Add lab evaluator and training reward API⁠

    Verified signature
    e38c605

  1. Update FlavourBench authorship and affiliation

    Verified signature
    854cc74
  2. Deploy final 27-model complete-core explorer⁠

    Verified signature
    1c950b5

  1. Publish v43 OpenRouter Fable dashboard⁠

    Verified signature
    170f6ba
  2. Clarify Fable refusal diagnostics

    Verified signature
    430599e
  3. Finish Fable 5 powered release

    Verified signature
    8507864
  4. Update 26-model FlavourBench explorer

    Verified signature
    5f69dee

  1. Add author affiliations⁠

    Verified signature
    dba402e
  2. Add FlavourBench co-authors⁠

    Verified signature
    a70a23e
  3. Publish the powered FlavourBench evidence explorer

    Verified signature
    91c562d

  1. Sharpen frontier benchmark positioning⁠

    Verified signature
    158edce
  2. Make leaderboard score-first with statistical intervals

    Verified signature
    a5a9440
  3. Position FlavourBench as frontier culinary benchmark

    Verified signature
    b2aac0d
  4. Add release upload hygiene

    Verified signature
    fb3ab5d
  5. Remove generated bytecode cache

    Verified signature
    a973596
  6. Retitle FlavourBench around executable evaluation

    Verified signature
    ced4adf
  7. Publish private FlavourBench explorer candidate

    Verified signature
    f590e74
  8. initial commit

    Verified signature
    6d1c70d