542generated behavioral tests
80.3%official best score
74.2%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 | 74.2% | ok | 402/542 | $4.72 | 111 | 0.26h | manifest · eval json · eval summary · usage audit |
| 2 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 71.0% | ok | 385/542 | n/a | 66 | 0.23h | not exported |
| 3 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 64.4% | ok | 349/542 | n/a | 96 | 0.31h | 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
74.2% from 402/542 ProgramBench-scored tests. Raw public eval statuses: failure: 188, passed: 575, skipped: 3.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch, exact version/output mismatch.
dc12d119a700eval.tests.test_caching.test_cache_file_flag_long (failure)dc12d119a700eval.tests.test_basic_invocation.test_version_short_flag (failure)dc12d119a700eval.tests.test_advanced_features.test_no_sqlite_writer_flag (failure)dc12d119a700eval.tests.test_basic_invocation.test_verbose_flag (failure)dc12d119a700eval.tests.test_caching.test_cache_file_flag_short (failure)dc12d119a700eval.tests.test_file_formats.test_avro_file_basic_query (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 | Claude Sonnet 4.6 | Anthropic | 80.3% | $32.92 | 610 |
| 2 | GPT 5.5 (xhigh) | OpenAI | 79.3% | $5.95 | 60 |
| 3 | Claude Opus 4.6 | Anthropic | 76.4% | $14.97 | 351 |
| 4 | Claude Opus 4.7 (xhigh) | Anthropic | 68.5% | $6.13 | 105 |
| 5 | Claude Opus 4.7 | Anthropic | 65.9% | $3.28 | 98 |
| 6 | GPT 5.5 (high) | OpenAI | 62.5% | $2.97 | 44 |
| 7 | GPT 5.5 | OpenAI | 54.2% | $1.27 | 23 |
| 8 | Gemini 3.1 Pro | 50.6% | $1.64 | 100 | |
| 9 | GPT 5.4 | OpenAI | 49.3% | $0.54 | 24 |
| 10 | Claude Haiku 4.5 | Anthropic | 30.3% | $0.77 | 134 |
| 11 | GPT 5 mini | OpenAI | 25.3% | $0.04 | 27 |
| 12 | GPT 5.4 mini | OpenAI | 10.1% | $0.03 | 8 |
| 13 | Gemini 3 Flash | 0.0% | $0.38 | 118 |