BlogLLM Benchmark Scores as Published, Side by Side
BixBench: Results as Published
Every result the catalog holds for BixBench: 5 models from 2 developers, read from 2 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 Science and MathMaintainer FutureHouse, in partnership with ScienceMachineIntroduced 2025Data version 2026103001, as of 2026-10-30
BixBench tests whether AI agents can do realistic bioinformatics analysis. The benchmark has 53 analysis scenarios built from published research notebooks, with 296 questions in total. An agent receives the data and must run the analysis and answer, and answers are graded either open-ended or as multiple choice.
BixBench
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.5OpenAI, released 2026-04-23 | 80.5% | OpenAI, Introducing GPT-5.5checked 2026-10-15 | Effort: xhigh |
| GPT-5.4OpenAI, released 2026-03-05 | 74% | OpenAI, Introducing GPT-5.5checked 2026-10-15 | Effort: xhigh |
| Grok 4.7xAI, released 2026-09-21 | 88.4% | xAI, Grok 4.7 model cardchecked 2026-10-14 | Effort: high |
| Grok 4.6xAI, released 2026-08-12 | 93.8% | xAI, Grok 4.7 model cardchecked 2026-10-14 | Effort: high |
| Grok 4.5xAI, released 2026-07-16 | 93.8% | xAI, Grok 4.7 model cardchecked 2026-10-14 | 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.