← Claude Sonnet 5 · API · High effort (default)
Evidenceacceptedvendor_reported
Terminal-Bench
Claude Sonnet 5 System Card (Terminal-Bench 2.1 / Terminus-2 harness)
ev_2026_07_sonnet5_terminal
Result
80.4%
Confidence B
- Tested
- 2026-06-30
- Runs
- 1
Configuration
How it was measured
- Exact model
- claude-sonnet-5
- Reasoning
- high
- Agent harness
- Terminus-2
- Tool access
- terminal / shell agent tools
- Prompting
- vendor system-card protocol
Caveats
Known limitations
- Vendor-reported on Terminus-2; confirm against public Terminal-Bench boards when available.
- Stretch evidence for STR / tool-calling, not a substitute for independent SWE boards.
Weighting
Evidence weight factors
The engine weights each evidence row by these factors (0–1) when fusing scores — higher is more trusted.
relevance0.90
quality0.55
recency1.00
health0.90
comparability0.80
reliability0.65
Benchmark
Terminal-Bench
Agentic terminal / tool-use suite measuring multi-step shell and environment task completion (Terminus-2 harness commonly cited).
Benchmark source →