1854generated behavioral tests
87.6%official best score
79.0%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 | 79.0% | ok | 1465/1854 | $5.41 | 87 | 0.23h | manifest · eval json · eval summary · usage audit |
| 2 | GPT 5.5 (xhigh) | 20260519T215046Z |
Codex /goal | Paper prompt + Goal contract no internet | 42.4% | ok | 786/1854 | n/a | 50 | 0.17h | not exported |
| 3 | GPT 5.5 (xhigh) | 20260518T232006Z |
Codex /goal | Paper prompt + /goal no internet | 42.3% | ok | 785/1854 | n/a | 44 | 0.16h | 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
79.0% from 1465/1854 ProgramBench-scored tests. Raw public eval statuses: failure: 476, passed: 1971.
Agent trace summary unavailable.
Likely miss class: behavioral mismatch, terminal rendering mismatch.
696c9c97d748tests.test_basic_invocation.test_completion_bash (failure)696c9c97d748tests.test_tui_advanced.test_tui_half_page_down (failure)696c9c97d748tests.test_tui_advanced.test_tui_screen_sync (failure)f61aaa36db33tests.test_config_detailed.test_force_screen_with_content (failure)f61aaa36db33tests.test_additional_main_coverage.test_argsToFiles_with_glob_pattern (failure)f61aaa36db33tests.test_coverage_target.test_exec_mode_simple_command (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 Opus 4.6 | Anthropic | 87.6% | $12.67 | 313 |
| 2 | Claude Sonnet 4.6 | Anthropic | 87.2% | $13.13 | 389 |
| 3 | GPT 5.5 (high) | OpenAI | 84.6% | $3.23 | 41 |
| 4 | GPT 5.5 (xhigh) | OpenAI | 79.8% | $10.99 | 88 |
| 5 | GPT 5.5 | OpenAI | 72.5% | $1.55 | 22 |
| 6 | Claude Opus 4.7 | Anthropic | 71.5% | $2.35 | 99 |
| 7 | Claude Opus 4.7 (xhigh) | Anthropic | 69.5% | $4.49 | 136 |
| 8 | Claude Haiku 4.5 | Anthropic | 57.2% | $0.48 | 97 |
| 9 | GPT 5.4 | OpenAI | 37.5% | $0.21 | 11 |
| 10 | GPT 5 mini | OpenAI | 36.2% | $0.01 | 10 |
| 11 | GPT 5.4 mini | OpenAI | 11.2% | $0.03 | 28 |
| 12 | Gemini 3 Flash | 4.1% | $0.19 | 62 | |
| 13 | Gemini 3.1 Pro | 2.0% | $2.37 | 111 |