2258generated behavioral tests
13.3%official best score
7.8%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) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 7.8% | ok | 175/2258 | n/a | 78 | 0.33h | not exported |
| 2 | GPT 5.5 (xhigh) | 20260517T094700Z |
Codex /goal | Mini-SWE-compatible no internet | 0.1% | ok | 3/2258 | $6.11 | 86 | 0.30h | manifest · eval json · eval summary · usage audit |
| 3 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 0.1% | ok | 3/2258 | n/a | 62 | 0.29h | 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) · 20260519T215046Z
Public eval evidence has not been exported for this row yet.
GPT 5.5 (xhigh) · 20260517T094700Z
0.1% from 3/2258 ProgramBench-scored tests. Raw public eval statuses: failure: 2357, passed: 108, skipped: 114.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch, exact version/output mismatch.
c0523186ac1eeval.tests.test_coverage.test_verilog_parsing (failure)c0523186ac1eeval.tests.test_coverage.test_gdscript_parsing (failure)c0523186ac1eeval.tests.test_coverage.test_ada_parsing (failure)c0523186ac1eeval.tests.test_coverage.test_objc_parsing (failure)c0523186ac1eeval.tests.test_coverage.test_bats_parsing (failure)c0523186ac1eeval.tests.test_coverage.test_org_parsing (failure)
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 | Claude Sonnet 4.6 | Anthropic | 13.3% | $58.28 | 659 |
| 2 | GPT 5.5 | OpenAI | 9.3% | $0.92 | 13 |
| 3 | Claude Opus 4.7 (xhigh) | Anthropic | 8.2% | $14.35 | 173 |
| 4 | GPT 5.5 (high) | OpenAI | 4.4% | $3.65 | 31 |
| 5 | GPT 5.4 | OpenAI | 3.6% | $0.20 | 8 |
| 6 | Gemini 3 Flash | 1.7% | $0.27 | 58 | |
| 7 | Gemini 3.1 Pro | 1.6% | $1.44 | 118 | |
| 8 | Claude Haiku 4.5 | Anthropic | 1.0% | $0.71 | 110 |
| 9 | GPT 5 mini | OpenAI | 0.9% | $0.03 | 16 |
| 10 | Claude Opus 4.7 | Anthropic | 0.5% | $0.29 | 22 |
| 11 | GPT 5.4 mini | OpenAI | 0.4% | $0.03 | 16 |
| 12 | GPT 5.5 (xhigh) | OpenAI | 0.0% | $3.75 | 34 |
| 13 | Claude Opus 4.6 | Anthropic | 0.0% | $2.49 | 226 |