BlogLLM Benchmark Scores as Published, Side by Side
Investment Banking Modeling Tasks (Internal): Results as Published
Every result the catalog holds for Investment Banking Modeling Tasks (Internal): 8 models from 1 developer, read from 3 pages. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Knowledge WorkMaintainer OpenAIIntroduced 2026Data version 2026103001, as of 2026-10-30
An internal OpenAI evaluation of investment banking modeling work (the financial models an analyst builds in a spreadsheet), listed with the professional evaluations on the GPT-5.5 release page. The page prints only the name and a pass rate per OpenAI model.
Investment Banking Modeling Tasks (Internal)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.5OpenAI, released 2026-04-23 | 88.5% | OpenAI, Introducing GPT-5.5checked 2026-10-15 | Effort: xhigh |
| GPT-5.5 ProOpenAI, released 2026-04-23 | 88.6% | OpenAI, Introducing GPT-5.5checked 2026-10-15 | Effort: xhigh |
| GPT-5.4OpenAI, released 2026-03-05 | 87.3% | OpenAI, Introducing GPT-5.4checked 2026-10-15; printed on 1 more page | Effort: xhigh |
| GPT-5.4 ProOpenAI, released 2026-03-05 | 83.6% | OpenAI, Introducing GPT-5.4checked 2026-10-15; printed on 1 more page | Effort: xhigh |
| GPT-5.3-CodexOpenAI, released 2026-02-05 | 79.3% | OpenAI, Introducing GPT-5.4checked 2026-10-15 | Effort: xhigh |
| GPT-5.2OpenAI, released 2025-12-11 | 68.4% | OpenAI, Introducing GPT-5.2checked 2026-10-15; printed on 1 more page | Effort: heavy |
| GPT-5.2 ProOpenAI, released 2025-12-11 | 71.7% | OpenAI, Introducing GPT-5.2checked 2026-10-15; printed on 1 more page | Effort: xhigh |
| GPT-5.1OpenAI, released 2025-11-12 | 59.1% | OpenAI, Introducing GPT-5.2checked 2026-10-15 | Effort: high |
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.