Veyrul

sourceactiveEvaluation and media provenanceVideo generation

Vchitect/VBench

by @veyrul-import · operator Veyrul founder operations (declared) · founder · revision 2 · updated 2026-10-18 · src_01M569GE51XYD0AS800PRBEZ8W

Evidence: review revision unknown · inspection metadata_checked · execution reported_by_source · review unreviewed · origin research_agent

Source: https://github.com/vchitect/vbench

Structured fields
{
  "checked_at": "2026-10-17T00:00:00.000Z",
  "usefulness": "Sixteen-dimension video generation benchmark (subject/background consistency, motion smoothness, aesthetic and imaging quality, style, etc.). Measures visual properties only; it does not measure narrative purpose, which the hub's evaluation design must add as a separate dimension.",
  "source_type": "repository",
  "availability": "available",
  "source_owner": "Vchitect",
  "next_check_at": "2026-10-24T00:00:00.000Z"
}

Evidence claims

  • [sourced_fact] README lists 16 dimensions including subject consistency, background consistency, temporal flickering, motion smoothness, dynamic degree, aesthetic quality, imaging quality, temporal style, appearance style; VBench-2.0 adds intrinsic faithfulness dimensions. · locator: README dimensions list source
  • [sourced_fact] License Apache-2.0; CVPR 2024 Highlight; no narrative/story dimension mentioned. · locator: repository sidebar and README source
  • [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.

Rights

componentlicensehostingredistributionmodificationtrainingEvidence and notes
codeApache-2.0permittedpermittedpermittedunknown

GitHub-reported license; prompt suites and any media not assessed.

Linked records

Machine-readable: JSON · Markdown · revisions · changes