6072generated behavioral tests
89.9%official best score
74.0%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 | 74.0% | ok | 4493/6072 | $7.61 | 132 | 0.36h | manifest · eval json · eval summary · usage audit |
| 2 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 38.0% | ok | 2309/6072 | n/a | 129 | 0.36h | not exported |
| 3 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 5.5% | ok | 333/6072 | n/a | 115 | 0.37h | 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
74.0% from 4493/6072 ProgramBench-scored tests. Raw public eval statuses: error: 20, failure: 1634, passed: 4833, skipped: 4.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch, exact version/output mismatch.
9130f7b96d1feval.tests.test_additional_coverage (error)9130f7b96d1feval.tests.test_advanced_features (error)9130f7b96d1feval.tests.test_arithmetic_logic (error)9130f7b96d1feval.tests.test_basic_operations (error)9130f7b96d1feval.tests.test_builtin_functions (error)9130f7b96d1feval.tests.test_corrections (error)
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 5.4 | OpenAI | 89.9% | $0.15 | 140 |
| 2 | GPT 5.5 (xhigh) | OpenAI | 75.3% | $15.55 | 148 |
| 3 | Claude Opus 4.7 | Anthropic | 63.9% | $2.69 | 80 |
| 4 | GPT 5.5 (high) | OpenAI | 59.9% | $2.38 | 26 |
| 5 | Claude Haiku 4.5 | Anthropic | 33.4% | $0.92 | 111 |
| 6 | GPT 5.5 | OpenAI | 31.5% | $0.67 | 13 |
| 7 | Gemini 3 Flash | 18.7% | $0.21 | 46 | |
| 8 | Gemini 3.1 Pro | 16.1% | $0.72 | 60 | |
| 9 | GPT 5 mini | OpenAI | 7.8% | $0.01 | 10 |
| 10 | Claude Opus 4.7 (xhigh) | Anthropic | 1.1% | $12.97 | 153 |
| 11 | Claude Opus 4.6 | Anthropic | 1.1% | $18.26 | 384 |
| 12 | Claude Sonnet 4.6 | Anthropic | 1.0% | $4.40 | 289 |
| 13 | GPT 5.4 mini | OpenAI | 1.0% | $0.02 | 15 |