Comparing 2 models
Gemini 3.5 Flash vs Gemini 3.8 Flash
Benchmarks through
Verdict
Gemini 3.8 Flash places higher on 11 of the 11 tasks where both are ranked. Gemini 3.8 Flash costs 1.8x less per input token.
- Choose Gemini 3.8 Flash for Creative Writing (#1), Translation (#1), Marketing Copy (#1).
Built from the published task rankings below; ranks come from each task's own weighted benchmarks, so a win on one task says nothing about another.
Measured head to head
Benchmarks both models were run on
| Evaluation | Gemini 3.5 Flash | Gemini 3.8 Flash | Source |
|---|---|---|---|
| Design Arena Coding | 1268 Elo | 1306 Elo | source · 2026-10-16 |
| Design Arena Full-stack Agent | 1197 Elo | 1244 Elo | source · 2026-10-14 |
| Design Arena UI | 1272 Elo | 1319 Elo | source · 2026-10-16 |
| Design Arena Web-app Agent | 1196 Elo | 1231 Elo | source · 2026-10-14 |
| Design Arena Website | 1269 Elo | 1303 Elo | source · 2026-10-16 |
| FACTS Grounding | 73.2% | 73.1% | source · 2026-10-16 |
| FrontierMath Tiers 1-3 (Epoch AI) | 62.8% ± 2.9 | 68.4% ± 2.8 | source · 2026-06-10 |
| GPQA Diamond (Epoch AI) | 88.9% ± 2.2 | 95.4% ± 1.4 | source · 2026-08-06 |
| GPQA Diamond (Epoch AI) | 86.4% ± 2.4 | 95.4% ± 1.4 | source · 2026-07-15 |
| GPQA Diamond (Epoch AI) | 92.8% ± 1.6 | 95.4% ± 1.4 | source · 2026-05-22 |
| LiveBench Agentic Coding | 49.0% | 54.2% | source · 2026-10-16 |
| LiveBench Coding | 78.2% | 72.5% | source · 2026-10-16 |
| LiveBench Data Analysis | 64.9% | 54.0% | source · 2026-10-16 |
| LiveBench Instruction Following | 75.6% | 81.4% | source · 2026-10-16 |
| LiveBench Language | 84.6% | 87.8% | source · 2026-10-16 |
| LiveBench Math | 88.2% | 91.6% | source · 2026-10-16 |
| LiveBench Reasoning | 82% | 89.3% | source · 2026-10-16 |
| LMArena Creative Writing | 1468 Elo | 1492 Elo | source · 2026-10-15 |
| LMArena Instruction Following | 1467 Elo | 1491 Elo | source · 2026-10-15 |
| LMArena Long Query | 1483 Elo | 1507 Elo | source · 2026-10-15 |
| LMArena Math | 1503 Elo | 1527 Elo | source · 2026-10-15 |
| LMArena Text | 1477 Elo | 1496 Elo | source · 2026-10-15 |
| LMArena Text Factuality | 1477 Elo | 1474 Elo | source · 2026-10-15 |
| LMArena Vision | 1284 Elo | 1289 Elo | source · 2026-10-09 |
| LMArena Vision Diagrams | 1309 Elo | 1333 Elo | source · 2026-10-09 |
| LMArena Vision OCR | 1293 Elo | 1317 Elo | source · 2026-10-09 |
| LMArena WebDev | 1499 Elo | 1585 Elo | source · 2026-10-13 |
| LMArena WebDev Image-to-Web | 1501 Elo | 1593 Elo | source · 2026-10-09 |
| OTIS Mock AIME (Epoch AI) | 88.9% ± 4.7 | 98.9% ± 0.9 | source · 2026-08-06 |
| OTIS Mock AIME (Epoch AI) | 80% ± 6 | 98.9% ± 0.9 | source · 2026-07-15 |
| OTIS Mock AIME (Epoch AI) | 95.6% ± 2.7 | 98.9% ± 0.9 | source · 2026-05-25 |
| SimpleQA Verified (Epoch AI) | 66.2% ± 1.5 | 69.7% ± 1.5 | source · 2026-08-27 |
Values are as published by each evaluator, in its own units; rows from different evaluators are not on a common scale.
task evidence
How they place when the question changes
Lower ranks are better. “Best” marks the strongest option in this shortlist for that task.