Evaluator Bench
tooldirectory.ai, 2026-07-08; retrieved 2026-09-14 tier 4 press (aggregator)confirmed

Andon Labs, agent safety evaluations

https://tooldirectory.ai/tools/andon-labs

Anthropic Project Vend partnership; results cited in model cards. Cited from a search snippet; not re-fetched in full. [verify 2026-09-15] fetched 2026-09-15; page loads, citation matches

Signals citing this source

  • Andon Labs, M (for): Benchmarks (Vending-Bench) cited in model cards across labs.“Vending-Bench results are cited in frontier-lab model cards and press”
  • Andon Labs, F (against): Commercial platform (Pion) and lab partnerships; funding and client mix not disclosed.
  • Andon Labs, A (against): Access through partnerships (Anthropic Project Vend); not routine pre-release.
  • Andon Labs, X (against): Operates AI-run vending and retail deployments inside Anthropic and xAI offices as commercial partnerships while benchmarking those labs' models.“offering custom evaluations and deployments to AI labs and agent builders”

Ledger rows citing this URL

  • T56 Project Vend partnership; AI-run vending deployed inside Anthropic offices confirmed