BlogLLM Benchmark Scores as Published, Side by Side
Impossible ExploitGym: Results as Published
Every result the catalog holds for Impossible ExploitGym: 1 model from 1 developer, read from 1 page. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Safety and AlignmentMaintainer OpenAI (internal evaluation, per the printed label)Data version 2026103001, as of 2026-10-30
Impossible ExploitGym is printed on the GPT-6 Astra system card. By its name alone it would be a variant of the public ExploitGym benchmark whose tasks cannot be solved as designed. No source found describes this variant, so its task set and scoring are unconfirmed.
Not checked against a maintainer's page.
Impossible ExploitGym
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-6 AstraOpenAI, released 2026-09-03 | 100% | OpenAI, GPT-6 Astrachecked 2026-10-14 | - |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.