2126generated behavioral tests
14.6%official best score
16.3%Codex best score
3Codex result rows
Codex Results by Model
| # | Model | Run | Agent | Mode | Score | Eval | Tests | Est. cost | Calls | Wall | Evidence |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 16.3% | ok | 347/2126 | n/a | 28 | 0.12h | not exported |
| 2 | GPT 5.5 (xhigh) | 20260517T094700Z |
Codex /goal | Mini-SWE-compatible no internet | 5.3% | ok | 112/2126 | $4.04 | 84 | 0.20h | manifest · eval json · eval summary · usage audit |
| 3 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 4.5% | ok | 96/2126 | n/a | 64 | 0.18h | not exported |
Why Scores Differ
Official ProgramBench rows are the public mini-SWE-agent submissions. GoalBench rows are separate Codex /goal submissions. Same model label does not mean the same agent, prompt, tool loop, or generated implementation. A GoalBench row is resolved only when the ProgramBench-scored pass rate is exactly 100%.
GPT 5.5 (xhigh) · 20260518T232006Z
Public eval evidence has not been exported for this row yet.
GPT 5.5 (xhigh) · 20260517T094700Z
5.3% from 112/2126 ProgramBench-scored tests. Raw public eval statuses: failure: 2179, passed: 187, skipped: 184.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch, exact version/output mismatch.
920f2c41e995eval.tests.test_additional_coverage.test_missingInclude_check (failure)920f2c41e995eval.tests.test_advanced_features.test_includes_file (failure)920f2c41e995eval.tests.test_additional_coverage.test_complex_pointer_error (failure)920f2c41e995eval.tests.test_additional_coverage.test_warning_exitcode (failure)920f2c41e995eval.tests.test_advanced_features.test_header_file_checking (failure)920f2c41e995eval.tests.test_advanced_features.test_library_with_check_library (failure)
GPT 5.5 (xhigh) · 20260519T215046Z
Public eval evidence has not been exported for this row yet.
Official ProgramBench Results by Model
| # | Model | Provider | Score | Cost | Calls |
|---|---|---|---|---|---|
| 1 | Claude Haiku 4.5 | Anthropic | 14.6% | $0.74 | 112 |
| 2 | GPT 5 mini | OpenAI | 13.8% | $0.04 | 15 |
| 3 | Claude Opus 4.6 | Anthropic | 10.6% | $6.95 | 175 |
| 4 | Gemini 3 Flash | 8.8% | $0.20 | 74 | |
| 5 | GPT 5.5 (xhigh) | OpenAI | 5.0% | $5.35 | 63 |
| 6 | Gemini 3.1 Pro | 4.6% | $1.64 | 94 | |
| 7 | Claude Opus 4.7 (xhigh) | Anthropic | 3.8% | $1.94 | 71 |
| 8 | Claude Sonnet 4.6 | Anthropic | 3.1% | $15.02 | 370 |
| 9 | Claude Opus 4.7 | Anthropic | 2.8% | $0.62 | 27 |
| 10 | GPT 5.4 | OpenAI | 2.4% | $0.24 | 9 |
| 11 | GPT 5.5 | OpenAI | 2.3% | $0.73 | 13 |
| 12 | GPT 5.5 (high) | OpenAI | 2.1% | $2.02 | 32 |
| 13 | GPT 5.4 mini | OpenAI | 1.9% | $0.03 | 11 |