ctxdexSynthetic model data for learning

Benchmark family

Terminal-Bench

A family groups the versions of one benchmark. It is not evaluated and has no applicable models of its own; each of its 5 definitions is resolved separately.

AgenticTerminal AgentTerminal

Definitions

Compare the newest 4 →

Newest first, following the recorded successor chain. Each definition is resolved against models separately.

  1. BenchmarkTerminal-Bench 4.04.0
    ActiveNewest in family
    Successor of Terminal-Bench 3.0
    Scale
    66 tasks
    Primary metric
    Accuracy
    Applicable models
    51
  2. BenchmarkTerminal-Bench 3.03.0
    Superseded
    Successor of Terminal-Bench 2.1
    Scale
    74 tasks
    Primary metric
    Accuracy
    Applicable models
    51
  3. BenchmarkTerminal-Bench 2.12.1
    Superseded
    Successor of Terminal-Bench 2.0
    Scale
    89 tasks
    Primary metric
    Accuracy
    Applicable models
    51
  4. BenchmarkTerminal-Bench 2.02.0
    Superseded
    Successor of Terminal-Bench 1.0
    Scale
    89 tasks
    Primary metric
    Accuracy
    Applicable models
    51
  5. BenchmarkTerminal-Bench 1.01.0
    Superseded
    Scale
    80 tasks
    Primary metric
    Accuracy
    Applicable models
    51
Related outside this family
Terminal-Bench-Science 0.1 — related to — Terminal-Bench 4.0

Family methodology

Versioning

Continuous semantic-versioned benchmark family. Material task/resource/evaluator changes cross release comparability boundaries and may require reruns.

Across definitions

Tasks
Terminal Use
Sources
None recorded at family level
Last reviewed
2026-09-23

Applicability does not mean evaluated; benchmark result/model score data is outside this UI contract.