BlogLLM Benchmark Scores as Published, Side by Side
$OneMillion-Bench: Results as Published
Every result the catalog holds for $OneMillion-Bench: 5 models from 3 developers, 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 Knowledge WorkMaintainer Paper authors from Humanlaya, BIGAI, xbench and M-A-PIntroduced 2026Data version 2026103001, as of 2026-10-30
$OneMillion-Bench tests language agents on 400 expert-curated professional tasks in law, finance, industry, healthcare and natural science. The tasks ask agents to find authoritative sources, resolve conflicting evidence, apply domain rules and make constrained decisions, and each task carries a dollar value based on the time a senior professional would need. Responses are scored with an expert rubric covering factual accuracy, logical coherence, practical feasibility and professional compliance.
$OneMillion-Bench (expert score)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.6 SolOpenAI, released 2026-07-09 | 53.8 | Alibaba, Qwen3.8-Max comparisonchecked 2026-10-14 | Effort: max |
| Claude Fable 5Anthropic, released 2026-06-09 | 55.9 | Alibaba, Qwen3.8-Max comparisonchecked 2026-10-14 | - |
| Claude Opus 4.8Anthropic, released 2026-05-28 | 41.8 | Alibaba, Qwen3.8-Max comparisonchecked 2026-10-14 | - |
| Qwen3.8-MaxAlibaba, released 2026-08-03 | 52.5 | Alibaba, Qwen3.8-Max comparisonchecked 2026-10-14 | - |
| Qwen3.7-MaxAlibaba | 44.4 | Alibaba, Qwen3.8-Max comparisonchecked 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.