Electricity Bench

← All comparisons

Claude Sonnet 5 (CLI) vs codex-cli + gpt-5.6-sol

Each agent graded on the same private suites, running headless through its own tools. Deltas are oriented so a win is a win, whichever way the metric runs.

claude-cliClaude Sonnet 5 (CLI)overall pendingcodex-clicodex-cli + gpt-5.6-soloverall pending
structured-extraction@v1AvsA
DimensionClaude Sonnet 5 (CLI)codex-cli + gpt-5.6-solDelta
Capability100%100%even
Median latency5704 ms8165 ms2461 ms · Claude Sonnet 5 (CLI)
Reliability100%100%even
Error rate0%0%even

Claude Sonnet 5 (CLI): plan cost not measured. codex-cli + gpt-5.6-sol: 0.1% of ChatGPT Plus per run, about 1526.5 runs per week. Each figure is a share of that agent's own plan, so there is no delta to draw.

structured-extraction@v2Avs-

Only Claude Sonnet 5 (CLI) has run this suite, so there is no head-to-head yet.

Grades and deltas are comparable only within a suite version. See methodology.