Gemini 3.5 Flash vs GPT-5.5
Gemini 3.5 Flash 
GPT-5.5 Verdict
- GPT-5.5 leads on quality: 66.6 vs 63.9.
- Gemini 3.5 Flash (medium) is stronger in coding, knowledge, multimodal.
- GPT-5.5 is stronger in agents & tools, composite, human preference, instruction following, long context, reasoning.
- Gemini 3.5 Flash (medium) is 3.3× cheaper ($3.38 vs $11.25 per 1M blended).
- Gemini 3.5 Flash (medium) streams 2.4× faster (209 vs 88 tokens per second).
| Metric | Gemini 3.5 Flash (medium) | GPT-5.5 |
|---|---|---|
| BenchLeader Index | 63.9 | 66.6 |
| Agents & tools score | 67.8 | 68.4 |
| Coding score | 58.9 | 57.3 |
| Composite score | 71.4 | 77.7 |
| Human preference score | 67.4 | 67.6 |
| Instruction following score | 72.2 | 73.3 |
| Knowledge score | 73.9 | 73.7 |
| Long context score | 63.6 | 68.8 |
| Multimodal score | 67.6 | 64.9 |
| Reasoning score | 66.8 | 69.7 |
| Blended price $/M | $3.38 | $11.25 |
| Output speed | 209 tok/s | 88 tok/s |
| Time to first answer | 17.0 s | 62.3 s |
| Context window | 1M | 1.1M |
| Terminal-Bench | – | 84.7% |
| SimpleBench | – | 69.0% |
| Remote Labor Index | – | 6.3% |
| APEX-Agents | – | 38.5% |
| FrontierCode | – | 43.0% |
| Epoch Capabilities Index | – | 159.1 |
| LMArena Text | 1476 | 1477 |
| LMArena Hard Prompts | 1493 | 1498 |
| LMArena Coding | 1503 | 1509 |
| LMArena WebDev | 1491 | 1458 |
| LMArena Vision | 1306 | 1296 |
| LMArena Agent | – | 3 |
| AA Intelligence Index | 33.6 | 38.6 |
| IFBench | 74.6% | 75.8% |
| AA-LCR | 74.3% | 84.3% |
| MMMU-Pro | 83.9% | 79.9% |
| AA-Omniscience | 20.8 | 20.5 |
| Terminal-Bench Hard | 39.4% | 60.6% |
| GPQA Diamond (AA) | 92.1% | 93.5% |
| Humanity's Last Exam (AA) | 41.3% | 45.8% |
| SciCode (AA) | – | 55.8% |
| τ²-Bench Telecom (AA) | 95.6% | 93.9% |
| HiL-Bench | – | 39.7% |
| EQ-Bench 4 | – | 1315 |
| Kagi LLM Benchmark | – | 88.8% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.