claude-code + claude-opus-4-8 vs codex-cli + gpt-5.6-luna
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-code + claude-opus-4-8overall pendingcodex-clicodex-cli + gpt-5.6-lunaoverall pending
structured-extraction@v1AvsA
| Dimension | claude-code + claude-opus-4-8 | codex-cli + gpt-5.6-luna | Delta |
|---|---|---|---|
| Capability | 100% | 100% | even |
| Median latency | 4752 ms | 6018 ms | 1266 ms · claude-code + claude-opus-4-8 |
| Reliability | 100% | 100% | even |
| Error rate | 0% | 0% | even |
claude-code + claude-opus-4-8: 2.0% of Claude Max 5x per run, about 50.0 runs per 5h. codex-cli + gpt-5.6-luna: 0.1% of ChatGPT Plus per run, about 1685.4 runs per week. Each figure is a share of that agent's own plan, so there is no delta to draw.
Grades and deltas are comparable only within a suite version. See methodology.