Gemini 3.8 Flash: real-world results
Gemini 3.8 Flash passed 34 of 35 runs (97%) on 35 different use cases.
Average cost per run was $0.35, average wall time 4m, with 0.17 human step-ins per run.
Results by category
| Category | Runs | Passed |
|---|---|---|
| Creative / Animated Scene (one HTML File) | 12 | 12/12 |
| Software & DevOps | 2 | 2/2 |
| Allotments / Lettings | 1 | 0/1 |
| Community Transport / Volunteer Drivers | 1 | 1/1 |
| Council / Parking Permits | 1 | 1/1 |
| Finance / Expense Claims | 1 | 1/1 |
| Hotel / Billing | 1 | 1/1 |
| Leisure Centre / Lesson Enrolments | 1 | 1/1 |
| Library / Overdue Fines | 1 | 1/1 |
| Marine / Anchoring | 1 | 1/1 |
| Marine / Collision Avoidance | 1 | 1/1 |
| Marine / Environmental Monitoring | 1 | 1/1 |
| Marine / Fishing Compliance | 1 | 1/1 |
| Marine / Fishing Planning | 1 | 1/1 |
| Marine / Navigation | 1 | 1/1 |
| Marine / Powerboat Planning | 1 | 1/1 |
| Marine / Sailing Instruments | 1 | 1/1 |
| Marine / Sailing Performance | 1 | 1/1 |
| Marine / Tides and Pilotage | 1 | 1/1 |
| Marine / Weather Routing | 1 | 1/1 |
| Payroll / Timesheets | 1 | 1/1 |
| Pharmacy / Home Deliveries | 1 | 1/1 |
| Small Business / Credit Control | 1 | 1/1 |
Head-to-head comparisons
- Gemini 3.8 Flash vs Claude Haiku 5.5
- Gemini 3.8 Flash vs Claude Opus 5.5
- Gemini 3.8 Flash vs Claude Sonnet 5.5
- Gemini 3.8 Flash vs DeepSeek V4.1 Flash (Venice)
- Gemini 3.8 Flash vs Gemini 4 Argon
- Gemini 3.8 Flash vs Gemma 4 31B Instruct (Venice)
- Gemini 3.8 Flash vs GPT-6.1 Sol
- Gemini 3.8 Flash vs Mistral Small 3.2 24B Instruct (Venice)
- Gemini 3.8 Flash vs Qwen3 Coder 480B Turbo (Venice)
- Gemini 3.8 Flash vs GLM 4.7 (Venice)
Browse every run by Gemini 3.8 Flash →
From published single-model runs only (all settings grouped). How we score.