BENCHMARK REGISTRY
Public Artificial Intelligence Benchmark Records Database
Search operators

brand: benchmark: record: model: metric: date: org:

Separate values with commas, or alternatives with uppercase OR. Example: brand:Meta, Anthropic benchmark:GPQA

Terminal-Bench: 2.0

Benchmark Family
Terminal-Bench
Version or Variant
2.0
Canonical Reference
terminal-bench/2-0
Release Date
Unknown
Status
ACTIVE

Table may be scrolled horizontally on narrow displays.

Record No. Model Benchmark Metric Score Eval Date Source Model ID
BR-002O47-005 Claude Opus 4.7 Terminal-Bench 2.0 Overall 69.4% Unknown Primary source ANTHR-O47
BR-00331P-004 Gemini 3.1 Pro Terminal-Bench 2.0 Overall 68.5% Unknown Primary source GOOGL-G31P
BR-005C2-003 Composer 2 Terminal-Bench 2.0 Overall 61.7 Unknown Primary source CURSR-C2
BR-00735-031 Qwen3.5-397B-A17B Terminal-Bench 2.0 Overall 52.5 Unknown Primary source QWEN-35
BR-008V4FLASH-014 DeepSeek V4-Flash Terminal-Bench 2.0 Overall 56.9 Unknown Primary source DPSK-V4FLASH
BR-008V4PRO-014 DeepSeek V4-Pro Terminal-Bench 2.0 Overall 67.9 Unknown Primary source DPSK-V4PRO