Finance AgentBrowse 284 families
Finance Agent
Can the model produce useful work in a professional knowledge-work setting?
Version / measure
- Domain
- Knowledge Work
- Ranked models
- 8
- Reported values
- 10
- Reports
- 2 · lab
- Direction
- Higher is better
- Status
- Version not specified
Rankings
Best reported score per model across effort settings
| # | Model | Best setting | Score |
|---|---|---|---|
| 1 | Gemini 3.5 FlashGoogle | reportedcost n/a1 value · May 19 · source | 57.9% |
| 2 | Muse Spark 1.1Meta | xhighcost n/a1 value · Jul 9 · source | 57.2% |
| 3 | Claude Opus 4.8Anthropic | maxcost n/a1 value · Jul 9 · source | 53.9% |
| 4 | GPT-5.5OpenAI | xhighcost n/a2 values · Jul 9 · source | 51.8% |
| 5 | Claude Opus 4.7Anthropic | reportedcost n/a1 value · May 19 · source | 51.5% |
| 6 | Claude Sonnet 4.6Anthropic | reportedcost n/a1 value · May 19 · source | 51.0% |
| 7 | Gemini 3.1 ProGoogle | highcost n/a2 values · Jul 9 · source | 43.0% |
| 8 | Gemini 3 FlashGoogle | reportedcost n/a1 value · May 19 · source | 42.6% |
Effort curve
Score vs. reported cost. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology
Can the model produce useful work in a professional knowledge-work setting?
Professional analysis, document, finance, office, or domain-specific tasks. Higher is better. The value is the percentage reported in this lab's table.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
927 expert-reviewed questions; values sourced from the cited evaluator.
Method / source ↗