Finance Agent
As of Nov 3, 2026, Finance Agent (v2.1) is maintained by Vals AI. Metric accuracy (%). Ceiling 100. Saturation unknown. Contamination risk low.
- Maintainer
- Vals AI
- Metric
- accuracy (%)
- Ceiling
- 100
- Saturation
- unknown
- Contamination
- low
- Cadence
- Declared cadence: as Vals updates (checked daily). Catalog snapshots: 1, latest Oct 31, 2026.
- Source
- Vals AI · this benchmark on Vals AI
- Grading
- AI judge (LLM-graded rubric) — Financial-analyst tasks graded against expert rubrics by a three-model LLM jury.
Source: Vals AI. A proprietary private test set run by Vals; scores are republished with permission. Rows are dated the day isitsota first saw them, and only the benchmark's standard harness counts.
Categories
Where it counts
- Finance — primary, weight 1
Top 5
Current top scores
- Claude Opus 5.5 · 67.846 · Oct 31, 2026
- Gemini 4 Argon · 67.658 · Oct 31, 2026
- Muse Wonder Full · 67.575 · Oct 31, 2026
- Claude Sonnet 5.5 · 67.275 · Oct 31, 2026
- Gemini 4 Carbon · 67.162 · Oct 31, 2026
| Model | Score | Date | Kind |
|---|---|---|---|
| Claude Opus 5.5 | 67.846 | 2026-10-31 | snapshot |
| Gemini 4 Argon | 67.658 | 2026-10-31 | snapshot |
| Muse Wonder Full | 67.575 | 2026-10-31 | snapshot |
| Claude Sonnet 5.5 | 67.275 | 2026-10-31 | snapshot |
| Gemini 3.8 Flash | 65.761 | 2026-10-31 | snapshot |
| Muse Wonder | 65.545 | 2026-10-31 | snapshot |
| DeepSeek V4.1 Flash | 65.428 | 2026-10-31 | snapshot |
| Claude Haiku 5.5 | 65.373 | 2026-10-31 | snapshot |
| GLM 5.3 | 65.004 | 2026-10-31 | snapshot |
| Mimo V2.6 Pro | 64.259 | 2026-10-31 | snapshot |
| Mimo V2.6 Flash | 63.316 | 2026-10-31 | snapshot |
| DeepSeek V4.1 Flash 1017 | 62.933 | 2026-10-31 | snapshot |
| GPT-6.1 Sol | 61.817 | 2026-10-31 | snapshot |
| Hy4 Preview | 61.623 | 2026-10-31 | snapshot |
| Mistral Large 4 | 61.383 | 2026-10-31 | snapshot |
| Muse Spark Open | 60.665 | 2026-10-31 | snapshot |
| GLM 5.3 Flash | 60.587 | 2026-10-31 | snapshot |
| Grok 4.7 | 60.585 | 2026-10-31 | snapshot |
| GPT-6 Astra | 60.428 | 2026-10-31 | snapshot |
| Kimi K3 | 60.013 | 2026-10-31 | snapshot |
| Step 5 Preview | 58.9 | 2026-10-31 | snapshot |
| MiniMax M3.1 Flash | 58.826 | 2026-10-31 | snapshot |
| DeepSeek V4 Pro 0813 | 57.116 | 2026-10-31 | snapshot |
| GPT-6 Luna | 56.841 | 2026-10-31 | snapshot |
| Minimax M3 | 54.338 | 2026-10-31 | snapshot |
| Inkling | 49.873 | 2026-10-31 | snapshot |
| Inkling Small | 44.776 | 2026-10-31 | snapshot |
| Mercury 3 | 54.708 | 2026-10-31 | snapshot |
| Kimi K3.1 Pro | 61.488 | 2026-10-31 | snapshot |
| GPT-6.1 Terra | 55.958 | 2026-10-31 | snapshot |