BenchFlow

§ Positioning

Developer-workflow-native benchmarking - built by someone who ran terminal infrastructure, with self-serve evaluation tooling for code and enterprise agents.

Best for · Teams benchmarking coding and enterprise workflow agents.

Team · Founded by Xiangyi Li (ex-Terminal, Red Hat).

Focus · Terminal and code-focused evaluation infrastructure for benchmarking agent performance in developer workflows.

Notable customers · AI engineering teams (per company)

• Code · Math

  Prime Intellect

  the open-source backbone of the category

• Code · Software Engineering

  Mechanize

  deep, narrow, elite

• Enterprise · Long Horizon

  Collinear

  enterprise long-horizon environments

← Browse all companies