1245generated behavioral tests
9.3%official best score
14.0%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) | 20260517T094700Z |
Codex /goal | Mini-SWE-compatible no internet | 14.0% | ok | 174/1245 | $3.13 | 43 | 0.15h | manifest · eval json · eval summary · usage audit |
| 2 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 10.4% | ok | 129/1245 | n/a | 49 | 0.17h | not exported |
| 3 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 3.9% | ok | 49/1245 | n/a | 49 | 0.17h | 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) · 20260517T094700Z
14.0% from 174/1245 ProgramBench-scored tests. Raw public eval statuses: failure: 1166, passed: 187, skipped: 29.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch.
bbe83d4d7c01tests.test_disre.test_time_selection_begin_beyond_trajectory (failure)bbe83d4d7c01tests.test_angle.test_angle_distribution_basic (failure)bbe83d4d7c01tests.test_angle.test_correlation_with_insufficient_frames (failure)bbe83d4d7c01tests.test_dipoles.test_dipoles_custom_parameters (failure)bbe83d4d7c01tests.test_convert_trj.test_convert_trj_empty_time_range (failure)bbe83d4d7c01tests.test_clustsize.test_multiple_frames_time_evolution (failure)
GPT 5.5 (xhigh) · 20260519T215046Z
Public eval evidence has not been exported for this row yet.
GPT 5.5 (xhigh) · 20260518T232006Z
Public eval evidence has not been exported for this row yet.
Official ProgramBench Results by Model
| # | Model | Provider | Score | Cost | Calls |
|---|---|---|---|---|---|
| 1 | Gemini 3.1 Pro | 9.3% | $1.82 | 100 | |
| 2 | GPT 5.5 (high) | OpenAI | 7.1% | $5.07 | 51 |
| 3 | Claude Sonnet 4.6 | Anthropic | 5.0% | $32.36 | 562 |
| 4 | Claude Opus 4.7 | Anthropic | 4.4% | $6.17 | 153 |
| 5 | Claude Opus 4.6 | Anthropic | 4.3% | $14.33 | 298 |
| 6 | GPT 5.5 | OpenAI | 4.3% | $0.80 | 11 |
| 7 | GPT 5.4 | OpenAI | 3.2% | $0.32 | 9 |
| 8 | GPT 5.5 (xhigh) | OpenAI | 3.1% | $7.86 | 62 |
| 9 | GPT 5.4 mini | OpenAI | 2.3% | $0.04 | 8 |
| 10 | Claude Haiku 4.5 | Anthropic | 1.4% | $1.04 | 97 |
| 11 | Claude Opus 4.7 (xhigh) | Anthropic | 1.1% | $7.41 | 151 |
| 12 | GPT 5 mini | OpenAI | 0.9% | $0.02 | 11 |
| 13 | Gemini 3 Flash | 0.0% | $0.34 | 71 |