Research · Benchmark profile
Measures multi-step financial research and evidence synthesis by an agent system.
Data verified 15 July 2026 · Methodology 1.2.0
Benchmark score on Finance Agent v2
Gemini 3.5 Flash leads at 57.9%, followed by Muse Spark 1.1 (57.2%) and GPT-5.5 (51.8%).
Visual analysis
Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.
4 ranked models · higher is better
| Model | Rank | Provider | Score |
|---|---|---|---|
| Gemini 3.5 Flash | #1 | 57.9% | |
| Muse Spark 1.1 | #2 | Meta | 57.2% |
| GPT-5.5 | #3 | OpenAI | 51.8% |
| Gemini 3.1 Pro Preview | #4 | 43% |
Shows how quickly scores separate down the field
| Model | Rank | Provider | Score |
|---|---|---|---|
| Gemini 3.5 Flash | #1 | 57.9% | |
| Muse Spark 1.1 | #2 | Meta | 57.2% |
| GPT-5.5 | #3 | OpenAI | 51.8% |
| Gemini 3.1 Pro Preview | #4 | 43% |
Average within the visible cohort · descriptive only
| Provider | Models | Average score |
|---|---|---|
| Meta | 1 | 57.2% |
| OpenAI | 1 | 51.8% |
| 2 | 50.5% |
One best score per model · higher is better
| Rank | Model | Provider | License | Score |
|---|---|---|---|---|
| #1 | Gemini 3.5 Flash gemini-3.5-flash | closed | 57.9% | |
| #2 | Muse Spark 1.1 muse-spark-1-1 | Meta | closed | 57.2% |
| #3 | GPT-5.5 gpt-5.5 | OpenAI | closed | 51.8% |
| #4 | Gemini 3.1 Pro Preview gemini-3.1-pro-preview | closed | 43% |
The top of this snapshot is led by Gemini 3.5 Flash at 57.9%; third place is 6.1 points behind. The top-4 spread is 14.9 points.
About Finance Agent v2
Measures multi-step financial research and evidence synthesis by an agent system. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
Measures multi-step financial research and evidence synthesis by an agent system.
Gemini 3.5 Flash by Google currently leads with 57.9%.
4 models in the LuminaBench cohort have a qualifying score on this benchmark.
No. This benchmark is display-only and does not enter the overall Lumina composite.
Related
Rolling · 22 results · Weighted 3.5%
SimpleQARolling · 2 results · Weighted 3.5%
PaperBenchRolling · 2 results · Weighted 3.5%
BenchLM Knowledge priorbench-align-v5.1 · 33 results · Weighted 1.6%
AA Long Context ReasoningRolling · 96 results · Weighted 1.5%
AA-OmniscienceRolling · 94 results · Weighted 1.5%
