609generated behavioral tests
38.4%official best score
37.6%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 | 37.6% | ok | 229/609 | $5.23 | 91 | 0.24h | manifest · eval json · eval summary · usage audit |
| 2 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 30.9% | ok | 188/609 | n/a | 97 | 0.28h | not exported |
| 3 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 29.4% | ok | 179/609 | n/a | 57 | 0.19h | 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
37.6% from 229/609 ProgramBench-scored tests. Raw public eval statuses: failure: 381, passed: 300.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch, exact version/output mismatch.
5695bac60382tests.test_basic.test_basic_png_conversion (failure)5695bac60382tests.test_basic.test_auto_generated_output_filename (failure)5695bac60382tests.test_basic.test_overwrite_protection_without_flag (failure)5695bac60382tests.test_basic.test_verbose_mode (failure)5695bac60382tests.test_basic.test_overwrite_with_flag (failure)5695bac60382tests.test_basic.test_complex_multi_box_diagram (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 | GPT 5.5 (xhigh) | OpenAI | 38.4% | $10.67 | 99 |
| 2 | GPT 5.5 (high) | OpenAI | 29.4% | $2.75 | 27 |
| 3 | Claude Opus 4.7 (xhigh) | Anthropic | 24.0% | $5.13 | 110 |
| 4 | Gemini 3 Flash | 20.4% | $0.19 | 53 | |
| 5 | GPT 5.5 | OpenAI | 19.2% | $1.27 | 16 |
| 6 | Claude Opus 4.6 | Anthropic | 18.4% | $14.43 | 260 |
| 7 | Claude Sonnet 4.6 | Anthropic | 16.9% | $38.31 | 461 |
| 8 | GPT 5.4 mini | OpenAI | 1.5% | $0.04 | 7 |
| 9 | Claude Haiku 4.5 | Anthropic | 0.8% | $0.38 | 90 |
| 10 | Claude Opus 4.7 | Anthropic | 0.7% | $2.54 | 76 |
| 11 | GPT 5.4 | OpenAI | 0.7% | $0.46 | 12 |
| 12 | Gemini 3.1 Pro | 0.7% | $0.89 | 44 | |
| 13 | GPT 5 mini | OpenAI | 0.7% | $0.01 | 10 |