Gemma 4 31B Instruct (Venice) vs GPT-6.1 Sol
On the 14 use cases both models ran, Gemma 4 31B Instruct (Venice) won 11 to 3 (better outcome first, then lower cost).
Pass rate: Gemma 4 31B Instruct (Venice) 79% (14 runs) vs GPT-6.1 Sol 100% (14 runs). Average cost per run: $0.0074 vs $0.05.
| Metric | Gemma 4 31B Instruct (Venice) | GPT-6.1 Sol |
|---|---|---|
| Runs on shared use cases | 14 | 14 |
| Pass rate | 79% (11/14) | 100% (14/14) |
| Avg cost per run | $0.0074 | $0.05 |
| Avg wall time | 9m | 1m |
| Step-ins per run | 0.00 | 0.00 |
| Use cases won | 11 | 3 |
Use case by use case
- Backtest a moving-average crossover on a bundled synthetic price file: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Best fishing windows from tide, solunar and weather data: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Best sailing angle and leg times from a polar table: GPT-6.1 Sol better (pass 0% vs 100%)
- Boat fuel range with a reserve over a planned trip: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Catch log checked against a made-up fishing regulations table: GPT-6.1 Sol better (pass 0% vs 100%)
- Discounted cash flow valuation with mid-year option and a sensitivity grid: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Go/no-go departure time from a gridded wind file: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Progressive tax on fictional brackets, and pre-tax vs after-tax savings comparison: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Read a synthetic annual-report excerpt and compute financial ratios: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Rebalance to target weights with whole shares and highest-cost-first tax lots: GPT-6.1 Sol better (pass 0% vs 100%)
- Required minimum distributions from a supplied divisor table, with and without delaying the first one: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Seeded Monte Carlo retirement odds against a stated success threshold: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Turn weekly ad-platform exports into an exact wasted-spend report: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
- Work backward from a retirement budget to a savings target, contributions and earliest retirement age: Gemma 4 31B Instruct (Venice) better (pass 100% vs 100%)
Only use cases both models ran are compared; all settings of each model are grouped. How we score.