855generated behavioral tests
97.2%official best score
95.1%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 | 95.1% | ok | 813/855 | $2.94 | 58 | 0.24h | manifest · eval json · eval summary · usage audit |
| 2 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 93.6% | ok | 800/855 | n/a | 52 | 0.15h | not exported |
| 3 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 43.0% | ok | 368/855 | n/a | 48 | 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) · 20260517T094700Z
95.1% from 813/855 ProgramBench-scored tests. Raw public eval statuses: failure: 43, passed: 840, skipped: 1.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch, exact version/output mismatch.
6192bcb4bfe7eval.tests.test_edge_cases.test_output_to_dash_means_stdout (failure)6192bcb4bfe7eval.tests.test_edge_cases.test_leading_delimiter (failure)6192bcb4bfe7eval.tests.test_header_selection.test_header_with_escaped_sequences (failure)6192bcb4bfe7eval.tests.test_output_handling.test_escaped_output_delimiter (failure)6b3e26672313eval.tests.test_hck.test_escaped_output_delimiter (failure)6b3e26672313eval.tests.test_hck.test_output_to_stdout_with_dash (failure)
GPT 5.5 (xhigh) · 20260518T232006Z
Public eval evidence has not been exported for this row yet.
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 | GPT 5.5 (xhigh) | OpenAI | 97.2% | $7.85 | 94 |
| 2 | GPT 5.5 (high) | OpenAI | 95.9% | $2.63 | 36 |
| 3 | Claude Opus 4.7 | Anthropic | 95.7% | $5.84 | 142 |
| 4 | Claude Opus 4.6 | Anthropic | 94.6% | $11.60 | 302 |
| 5 | Claude Opus 4.7 (xhigh) | Anthropic | 94.4% | $11.23 | 119 |
| 6 | GPT 5.5 | OpenAI | 91.8% | $0.73 | 13 |
| 7 | Claude Sonnet 4.6 | Anthropic | 87.6% | $15.93 | 374 |
| 8 | GPT 5.4 | OpenAI | 86.8% | $0.25 | 10 |
| 9 | GPT 5 mini | OpenAI | 80.5% | $0.03 | 13 |
| 10 | GPT 5.4 mini | OpenAI | 69.5% | $0.03 | 9 |
| 11 | Claude Haiku 4.5 | Anthropic | 64.9% | $0.57 | 107 |
| 12 | Gemini 3.1 Pro | 58.7% | $1.87 | 129 | |
| 13 | Gemini 3 Flash | 3.3% | $0.33 | 114 |