Gemini 3.5 Flash vs Gemini 3.8 Flash
Gemini 3.5 Flash Verdict
- Gemini 3.5 Flash (medium) and Gemini 3.8 Flash (medium) are level on quality (63.9 vs 63.5).
- Gemini 3.5 Flash (medium) is stronger in agents & tools, human preference, instruction following, reasoning.
- Gemini 3.8 Flash (medium) is stronger in coding, composite, knowledge, long context, multimodal.
- Gemini 3.8 Flash (medium) is 2.3× cheaper ($1.50 vs $3.38 per 1M blended).
- Gemini 3.8 Flash (medium) streams 1.3× faster (273 vs 209 tokens per second).
| Metric | Gemini 3.5 Flash (medium) | Gemini 3.8 Flash (medium) |
|---|---|---|
| BenchLeader Index | 63.9 | 63.5 |
| Agents & tools score | 67.8 | – |
| Coding score | 58.9 | 63.6 |
| Composite score | 71.4 | 79.4 |
| Human preference score | 67.4 | – |
| Instruction following score | 72.2 | – |
| Knowledge score | 73.9 | 77.6 |
| Long context score | 63.6 | 68.7 |
| Multimodal score | 67.6 | 68.9 |
| Reasoning score | 66.8 | – |
| Blended price $/M | $3.38 | $1.50 |
| Output speed | 209 tok/s | 273 tok/s |
| Time to first answer | 17.0 s | 13.8 s |
| Context window | 1M | 1M |
| SciCode | – | 54.4% |
| LMArena Text | 1476 | – |
| LMArena Hard Prompts | 1493 | – |
| LMArena Coding | 1503 | – |
| LMArena WebDev | 1491 | – |
| LMArena Vision | 1306 | – |
| AA Intelligence Index | 33.6 | 40 |
| IFBench | 74.6% | – |
| AA-LCR | 74.3% | 84.0% |
| MMMU-Pro | 83.9% | 84.2% |
| AA-Omniscience | 20.8 | 28.6 |
| Terminal-Bench Hard | 39.4% | – |
| GPQA Diamond (AA) | 92.1% | 93.5% |
| Humanity's Last Exam (AA) | 41.3% | 42.1% |
| SciCode (AA) | – | 55.1% |
| τ²-Bench Telecom (AA) | 95.6% | – |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.