BlogLLM Benchmark Scores as Published, Side by Side
FrontierFinance: Results as Published
Every result the catalog holds for FrontierFinance: 6 models from 5 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 Samaya AIIntroduced 2026Data version 2026103001, as of 2026-10-30
FrontierFinance tests finance agents across the investor workflow, from screening and discovery through research and evaluation to monitoring. It has 220 expert-written queries graded against 11,543 source-attributed rubrics. A response is scored by the share of expert rubrics it satisfies, macro-averaged across queries, with three judge models voting by majority.
FrontierFinance
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-6 AstraOpenAI, released 2026-09-03 | 45% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Finance Agent v2; Fallback: with fallback |
| Claude Opus 5.5Anthropic, released 2026-09-22 | 56.7% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Finance Agent v2; Fallback: with fallback |
| Claude Fable 5.1Anthropic, released 2026-09-01 | 60.8% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Finance Agent v2; Fallback: with fallback |
| DeepSeek-V4.1-Flash-1017DeepSeek, released 2026-10-17 | 54.5% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Finance Agent v2 |
| Kimi K3.1 ProMoonshot AI, released 2026-10-23 | 58.6% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Finance Agent v2 |
| GLM-5.3Z.ai, released 2026-08-18 | 57.5% | Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24 | Effort: max; Harness: Finance Agent v2 |
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.