| Benchmark | Gemini 3.5 Flash | Qwen3.6 Plus |
|---|---|---|
| AA Agentic Index | 70.4 | 29 |
| AA Intelligence | 52 | 40.5 |
| AA-LCR | 81 | 72.3 |
| AA-Omniscience | 21.2 | 2.6 |
| aime_2026 | 95 | 95.3 |
| arena_elo | 1477 | 1444 |
| arena_text_factuality | 1466 | 1438 |
| Artificial Analysis Coding Index | 70.1 | 54.5 |
| coding_arena_elo | 1490 | 1460 |
| critpt | 13.1 | 2.9 |
| Finance Agent v2 | 57.9 | 40.9 |
| frontiermath_tier_4 | 14.6 | 8.3 |
| gdpval | 57.8 | 32 |
| GPQA Diamond | 92.2 | 90.4 |
| HLE | 42.7 | 28.8 |
| hmmt_feb_2026 | 95.5 | 87.8 |
| IFBench | 76.3 | 75.2 |
| livebench | 75 | 70.8 |
| livebench_agentic_coding | 49 | 41.4 |
| livebench_coding | 78.2 | 78.2 |
| livebench_data_analysis | 64.9 | 69.9 |
| livebench_instruction_following | 75.6 | 58.3 |
| livebench_language | 84.6 | 75 |
| livebench_math | 88.2 | 83.7 |
| livebench_reasoning | 82 | 75.8 |
| MCP Atlas | 83.6 | 74.1 |
| MMMU-Pro | 84.3 | 78.8 |
| OmniScience Accuracy | 51.4 | 26.4 |
| OmniScience Non-Hallucination | 38.2 | 68 |
| OSWorld-Verified | 78.4 | 62.5 |
| scicode | 53.1 | 40.7 |
| simpleqa_verified | 68.4 | 49.1 |
| SWE-bench Pro | 55.1 | 56.6 |
| TauBench V3 - Banking | 32.2 | 20.8 |
| Terminal-Bench 2.0 | 76.2 | 61.6 |
| Terminal-Bench 2.1 | 78.7 | 61.4 |
| Terminal-Bench Hard | 46.2 | 43.9 |
| toolathlon | 56.5 | 39.8 |
| τ²-Bench Telecom (AA run) | 95.6 | 97.7 |
| τ³-Bench | 25.4 | 70.7 |
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.