2047generated behavioral tests
75.7%official best score
8.6%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) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 8.6% | ok | 176/2047 | n/a | 47 | 0.13h | not exported |
| 2 | GPT 5.5 (xhigh) | 20260517T094700Z |
Codex /goal | Mini-SWE-compatible no internet | 4.0% | ok | 82/2047 | $6.90 | 118 | 0.31h | manifest · eval json · eval summary · usage audit |
| 3 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 4.0% | ok | 82/2047 | n/a | 53 | 0.15h | 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) · 20260518T232006Z
Public eval evidence has not been exported for this row yet.
GPT 5.5 (xhigh) · 20260517T094700Z
4.0% from 82/2047 ProgramBench-scored tests. Raw public eval statuses: error: 31, failure: 2908, passed: 203, skipped: 14.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch.
e4def75fb3eftests.test_additional_stdlib.test_some_false (failure)e4def75fb3eftests.test_comprehensive_json.test_streaming_json_empty_arrays (failure)e4def75fb3eftests.test_comprehensive_json.test_json_object_single_property (failure)e4def75fb3eftests.test_advanced_operations.test_object_assign (failure)e4def75fb3eftests.test_coverage_boost.test_string_charcodeat (failure)e4def75fb3eftests.test_builtin_functions.test_exit_function_one (failure)
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 | Claude Opus 4.6 | Anthropic | 75.7% | $20.74 | 396 |
| 2 | GPT 5.5 (high) | OpenAI | 74.4% | $3.29 | 50 |
| 3 | GPT 5.5 | OpenAI | 56.5% | $1.21 | 24 |
| 4 | Claude Opus 4.7 | Anthropic | 49.3% | $5.93 | 145 |
| 5 | GPT 5.4 | OpenAI | 49.3% | $0.20 | 10 |
| 6 | Claude Opus 4.7 (xhigh) | Anthropic | 47.0% | $7.48 | 135 |
| 7 | GPT 5.5 (xhigh) | OpenAI | 43.3% | $7.63 | 71 |
| 8 | Gemini 3.1 Pro | 42.1% | $1.37 | 127 | |
| 9 | GPT 5.4 mini | OpenAI | 40.3% | $0.03 | 7 |
| 10 | Claude Haiku 4.5 | Anthropic | 17.5% | $0.73 | 133 |
| 11 | Gemini 3 Flash | 16.1% | $0.19 | 85 | |
| 12 | GPT 5 mini | OpenAI | 11.6% | $0.01 | 11 |
| 13 | Claude Sonnet 4.6 | Anthropic | 4.3% | $25.98 | 621 |