Gemini 3.8 Flash vs Gemma 4 31B
| Spec | Gemini 3.8 Flash | Gemma 4 31B |
|---|---|---|
| Respan Index | 54.5 (#31) | 18.7 (#140) |
| Blended price per 1M | $1.5 | - |
| Context window | 1.0M | 256K |
| Input / output per 1M | $0.75 / $3.75 | $0 / $0 |
| Max output | 66K | - |
| Input | text, vision, audio | text, vision |
| Released | 2026-09-02 | 2026-04-02 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | Gemini 3.8 Flash | Gemma 4 31B |
|---|---|---|
| Coding | 36.1 Gemini 3.8 Flash | 7.8 Gemma 4 31B |
| Agents | 66.7 Gemini 3.8 Flash | 12.5* Gemma 4 31B |
| Reasoning | 53.4 Gemini 3.8 Flash | 19.7 Gemma 4 31B |
| Knowledge & language | 70.9 Gemini 3.8 Flash | 11.0 Gemma 4 31B |
| Human preference | 56.0 Gemini 3.8 Flash | 50.4 Gemma 4 31B |
* Estimated: no published scores in that category. How the Respan Index works
Benchmarks both report
Gemini 3.8 Flash leads on 9 of the 10 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Gemini 3.8 Flash | Gemma 4 31B |
|---|---|---|
| LMArena Elo | 1497 | 1452 |
| LMArena Vision (LMArena) | 1293 | 1265 |
| LMArena WebDev (LMArena) | 1584 | 1370 |
| SAGE 1.1 (Vals AI) | 88.2% | 82.8% |
| Chess Puzzles (Epoch AI) | 58% | 5% |
| GPQA Diamond | 95.4% | 84.3% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.9% | 73.3% |
| SAGE (Vals AI) | 35.1% | 55% |
| SimpleQA Verified | 69.7% | 10.4% |
| WeirdML (Håvard Tveit Ihle) | 84.8% | 52.3% |
FAQ
Is Gemini 3.8 Flash better than Gemma 4 31B?
Gemini 3.8 Flash scores 54.5 on the Respan Index (#31), ahead of Gemma 4 31B at 18.7 (#140). Gemini 3.8 Flash leads on 9 of the 10 benchmarks both report.
Which has the larger context window?
Gemini 3.8 Flash accepts 1.0M tokens and Gemma 4 31B 256K tokens.