Is gemini-3.5-flash-high good at Data & Analytics?
gemini-3.5-flash-high ranks #19 of 69 for Data & Analytics — excellent. The top pick for this task is claude-opus-4.8-low.
gemini-3.5-flash-high on each Data & Analytics sub-task
| SQL Reasoning | 100.0/100 | #19 |
| Spot the Misleading Stat | 100.0/100 | #38 |
| Metric Calculation | 98.8/100 | #18 |
| Honest Communication | 96.5/100 | #24 |
Real examples, graded
WeakSurvivorship bias (Northwind) 55/100
“The model correctly identifies correlation vs. causation and reverse causality, and provides excellent business recommendations. However, it completely misses survivorship bias (selecting on the dependent variable), which is a critical flaw in the executive's reasoning and an explicitly stated failure mode for this task. It fails to point out that the executive only looked at successful accounts and ignored unsuccessful or churned accounts.”
WeakRelative vs absolute risk (Lumen) 26/100
“While the model correctly calculates and reports both the absolute (1 percentage point) and relative (50%) reductions, its core conclusion completely fails the benchmark. It explicitly treats an observational association as proven causation ('halves', 'prevent one adverse event', 'mitigating risk') and fabricates statistical significance, violating the core rules of honest data communication.”
Frequently asked
Is gemini-3.5-flash-high good at Data & Analytics?
gemini-3.5-flash-high ranks #19 of 69 models we tested for Data & Analytics, scoring excellent.
What is gemini-3.5-flash-high's strongest Data & Analytics skill?
Its best sub-task here is SQL Reasoning.
This page is Spring Prompt, running
We just did this for every model. Do it for your prompt.
The rankings above come from running real tasks through real models and scoring every output. Spring Prompt is that same engine — pointed at your prompt, your test cases, and your definition of good.
- Generate test cases from your prompt — no eval set required to start.
- Compare models side by side with quality, cost and latency in one matrix.
- Optimise the winner until the scores say it's ready to ship.
Prompt × model results
12 test cases · 3 evals