| Benchmark | Gemini 3 Flash Preview | Qwen3.7 Plus Preview |
|---|---|---|
| AA Agentic Index | 49.7 | 20.8 |
| AA Intelligence | 38.7 | 39.4 |
| AA-LCR | 73 | 69 |
| AA-Omniscience | 10.1 | 2.4 |
| apexAgents | 27.7 | 22.4 |
| arena_elo | 1473 | 1458 |
| arena_text_factuality | 1452 | 1454 |
| arena_vision | 1260 | 1263 |
| Artificial Analysis Coding Index | 42.6 | 55.9 |
| charxiv_rq | 80.3 | 85.9 |
| critpt | 8.6 | 9.1 |
| Finance Agent v2 | 42.6 | 38.2 |
| gdpval | 35.2 | 22.2 |
| GPQA Diamond | 90.4 | 90.3 |
| HLE | 43.5 | 35.6 |
| hmmt_feb_2026 | 89.4 | 92.9 |
| IFBench | 78 | 79.1 |
| MCP Atlas | 62 | 73.2 |
| mmmlu | 91.8 | 89 |
| MMMU-Pro | 81.2 | 80.5 |
| OmniDocBench 1.5 | 12.1 | 91.4 |
| OmniScience Accuracy | 53.4 | 22.5 |
| OmniScience Non-Hallucination | 7.6 | 74.5 |
| OSWorld-Verified | 65.1 | 73.3 |
| scicode | 50.6 | 51.3 |
| screenspot_pro_no_tools | 69.1 | 79 |
| SWE-bench Multilingual | 72.7 | 75.8 |
| SWE-bench Pro | 49.6 | 57.6 |
| SWE-bench Verified | 78 | 77.7 |
| TauBench V3 - Banking | 20.8 | 17.9 |
| Terminal-Bench 2.0 | 47.6 | 70.3 |
| Terminal-Bench 2.1 | 58 | 61 |
| Terminal-Bench Hard | 38.6 | 47 |
| VideoMMMU | 86.9 | 85.4 |
| τ²-Bench Telecom (AA run) | 80.4 | 93 |
| τ³-Bench | 17.5 | 17.9 |
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.