codex-cli + gpt-5.6-sol
codex-cli harness · ChatGPT Plus
operator ChatGPT Plus login shared between eb-runner and the claude host; graded runs only inside the scheduled quiet window, all recurring tasks blocked for its duration
Tier 1: localized single-file bug · tier 3: cause in a different module than the symptom · tier 5: problems that took a human hours. A cleared rung solved every task on it; a partial rung solved some.
overall pending Two distinct suite families are required for an overall grade.
Per-suite results
Run on codex-cli codex-cli 0.146.0, model gpt-5.6-sol (-c model_reasoning_effort=medium), Aug 4, 2026.
One run of this suite consumes 3.6% of ChatGPT Plus - about 27.8 runs per week. Measured per suite run (estimated-from-telemetry); an absolute cost on this plan, not comparable with another agent's plan. Measured in a declared quiet window on a shared subscription: operator ChatGPT Plus login shared between eb-runner and the claude host; graded runs only inside the scheduled quiet window, all recurring tasks blocked for its duration.
Trajectory
Every scored run of this agent, oldest first. Grades compare only within a suite version, so each version gets its own line; while a suite is being retired both versions run, and the superseded one is dashed. The filled point is the run the scorecard shows today. All agents over time →
Aug 4, 2026: F on v3 · Aug 4, 2026: D on v3
Compare
- codex-cli + gpt-5.6-sol vs antigravity-cli + gemini-3.1-pro
- codex-cli + gpt-5.6-sol vs antigravity-cli + gemini-3.6-flash
- codex-cli + gpt-5.6-sol vs claude-code + claude-fable-5
- codex-cli + gpt-5.6-sol vs claude-code + claude-haiku-4-5
- codex-cli + gpt-5.6-sol vs claude-code + claude-opus-5
- codex-cli + gpt-5.6-sol vs claude-code + claude-sonnet-5
- codex-cli + gpt-5.6-sol vs codex-cli + gpt-5.6-luna
- codex-cli + gpt-5.6-sol vs cursor-cli + composer-2.5 (Cursor Pro)
- codex-cli + gpt-5.6-sol vs grok-cli + grok-4.5
Transcript excerpts
Outputs are shown only when they can be separated safely from the private suite. Scores and run metadata remain visible when an output is withheld.
| Task | # | Turns | Latency | Output |
|---|---|---|---|---|
| b2r5h8 | 1 | 1 | 211436 ms | withheldPrivate-suite content |
| c8g2m4 | 1 | 1 | 104079 ms | withheldPrivate-suite content |
| d9b4x6 | 1 | 1 | 169342 ms | withheldPrivate-suite content |
| h5n2v7 | 1 | 1 | 404805 ms | withheldPrivate-suite content |
| h7w3j5 | 1 | 1 | 94713 ms | withheldPrivate-suite content |
| k4f9t7 | 1 | 1 | 289116 ms | withheldPrivate-suite content |
| m6t3q8 | 1 | 1 | 239320 ms | withheldPrivate-suite content |
| p1x9k4 | 1 | 1 | 238797 ms | withheldPrivate-suite content |
| q8k3j6 | 1 | 1 | 367637 ms | withheldPrivate-suite content |
| r3v8m5 | 1 | 1 | 545889 ms | withheldPrivate-suite content |
| s2j7f4 | 1 | 1 | 105861 ms | withheldPrivate-suite content |
| t5s8n2 | 1 | 1 | 83417 ms | withheldPrivate-suite content |
| w4j7q2 | 1 | 1 | 446230 ms | withheldPrivate-suite content |
| y3p7k1 | 1 | 1 | 177957 ms | withheldPrivate-suite content |
| z6q1v9 | 1 | 1 | 106785 ms | withheldPrivate-suite content |