Claude Sonnet 5.5 vs Gemini 3.6 Flash
Claude Sonnet 5.5 
Gemini 3.6 Flash Verdict
- Claude Sonnet 5.5 (max) leads on quality: 67.6 vs 60.4.
- Claude Sonnet 5.5 (max) is stronger in agents & tools, coding, composite, long context, maths, reasoning.
- Gemini 3.6 Flash (high) is stronger in knowledge, human preference, multimodal.
- Gemini 3.6 Flash (high) is 2.7× cheaper ($1.50 vs $4.00 per 1M blended).
- Gemini 3.6 Flash (high) streams 1.4× faster (197 vs 141 tokens per second).
| Metric | Claude Sonnet 5.5 (max) | Gemini 3.6 Flash (high) |
|---|---|---|
| BenchLeader Index | 67.6 | 60.4 |
| Agents & tools score | 69.5 | 52.0 |
| Coding score | 69.8 | 57.2 |
| Composite score | 72.6 | 57.0 |
| Knowledge score | 65.4 | 65.7 |
| Long context score | 66.6 | 65.2 |
| Maths score | 72.9 | 60.0 |
| Reasoning score | 81.1 | 62.8 |
| Human preference score | – | 67.9 |
| Multimodal score | – | 65.5 |
| Blended price $/M | $4.00 | $1.50 |
| Output speed | 141 tok/s | 197 tok/s |
| Time to first answer | 487.1 s | 13.7 s |
| Context window | 1M | 1.0M |
| GPQA Diamond | 95.6% | 94.1% |
| FrontierMath Tiers 1–3 | 88.8% | 59.0% |
| FrontierMath Tier 4 | 80.5% | 21.9% |
| OTIS Mock AIME | 100.0% | 94.2% |
| SimpleQA Verified | 46.5% | 66.2% |
| Terminal-Bench | 61.8% | – |
| SciCode | 61.0% | 52.7% |
| WeirdML | – | 56.1% |
| APEX-Agents | 75.5% | – |
| ProofBench | 100.0% | – |
| LMArena Text | – | 1482 |
| LMArena Hard Prompts | – | 1501 |
| LMArena Coding | – | 1518 |
| LMArena WebDev | – | 1538 |
| LMArena Vision | – | 1298 |
| LMArena Agent | 12 | -7.8 |
| LiveBench | 75.7% | 73.6% |
| LiveBench Reasoning | 91.6% | 85.2% |
| LiveBench Coding | 91.4% | 77.9% |
| LiveBench Agentic Coding | 56.3% | 43.4% |
| LiveBench Mathematics | 96.1% | 86.4% |
| LiveBench Data Analysis | 59.5% | 63.0% |
| LiveBench Language | 78.0% | 83.9% |
| LiveBench Instruction Following | 56.8% | 75.4% |
| AA Intelligence Index v4.3.2 | 56 | 34.0 |
| AA-LCR | 82.7% | 80.0% |
| MMMU-Pro | – | 83.2% |
| AA-Omniscience | 32.3 | 22.1 |
| GPQA Diamond (AA) | – | 92.8% |
| Humanity's Last Exam (AA) | 55.0% | 40.8% |
| SciCode (AA) | 61.0% | 53.4% |
| LiveCodeBench | – | 88.1% |
| MMLU-Pro | – | 89.3% |
| IOI | – | 35.1% |
| LegalBench | – | 86.7% |
| CorpFin | – | 63.3% |
| TaxEval | – | 74.9% |
| Terminal-Bench 2.1 (Vals) | – | 73.8% |
| SWE-bench (Vals) | – | 79.6% |
| GPQA Diamond (Vals) | – | 93.4% |
| ARC-AGI-1 | – | 91.2% |
| ARC-AGI-2 | – | 60.4% |
| CritPt | 31.4% | 10.6% |
| GDPval-AA v2.1 | 67.0% | 39.1% |
| τ³-Banking (AA) | – | 29.9% |
| Analyst Agent (AA) | 57.5% | – |
| BioMysteryBench | – | 58.5% |
| Code Migration | – | 30.9% |
| CyberBench | – | 45.4% |
| Excel Modeling Benchmark | – | 65.4% |
| Finance Agent v2 | – | 56.3% |
| Harvey's Legal Agent Benchmark | – | 3.3% |
| Legal Research Bench | – | 25.0% |
| MedCode | – | 53.1% |
| MedScribe | – | 79.7% |
| MMMU-Pro (Vals) | – | 88.4% |
| MortgageTax | – | 67.6% |
| ProgramBench | – | 0.0% |
| Public Benefits Bench | – | 56.6% |
| SAGE | – | 48.7% |
| Tax Agent Bench | – | 49.9% |
| Terminal-Bench Science | – | 4.3% |
| Vals Multimodal Index | – | 65.1% |
| Vibe Code Bench v1.1 | – | 64.0% |
| LMArena Maths | – | 1503 |
| LMArena Creative Writing | – | 1471 |
| LMArena Instruction Following | – | 1472 |
| LMArena Multi-turn | – | 1487 |
| LMArena Longer Queries | – | 1485 |
| LMArena Document | – | 1453 |
| Chess Puzzles | – | 40.0% |
| Mystery Game Puzzles | 65.0% | 30.0% |
| DeepSWE v1.1 | – | 46.7% |
| LMCA | – | 44.9% |
| DTBench | – | 95.5% |
| CursorBench | 55.5% | – |
| ALE-Bench | – | 715.5 |
| GDP.pdf | – | 14.0% |
| FrontierSWE | 61.9% | – |
| Terminal-Bench 4.0 (AA) | 63.6% | 7.1% |
| Terminal-Bench 2.1 (AA) | – | 77.5% |
| AutomationBench | 71.8% | – |
| GDP.pdf | 25.8% | – |
| MLCR | 75.0% | – |
| Harvey LAB | 2.8% | – |
| AA-Omniscience: accuracy | 54.0% | 50.0% |
| AA-Omniscience: non-hallucination | 53.0% | 44.4% |
| AA-Briefcase v1.1 | 1823 | – |
Data as of 2026-10-11. Best configuration of each model; every score links to its source on the model pages.
Claude Sonnet 5.5 vs Gemini 3.6 Flash: questions
- Is Claude Sonnet 5.5 better than Gemini 3.6 Flash?
- Claude Sonnet 5.5 (max) leads on quality: 67.6 vs 60.4. The BenchLeader Index combines every independent quality benchmark; Claude Sonnet 5.5 (max) is ahead overall as of 2026-10-11, but check the category scores for your use.
- Is Claude Sonnet 5.5 better than Gemini 3.6 Flash for coding?
- Claude Sonnet 5.5 scores higher in coding (70 vs 57 on the category index, where 50 is average).
- Is Claude Sonnet 5.5 better than Gemini 3.6 Flash for agentic tasks?
- Claude Sonnet 5.5 scores higher in agentic tasks (70 vs 52 on the category index, where 50 is average).
- Which is cheaper, Claude Sonnet 5.5 or Gemini 3.6 Flash?
- Gemini 3.6 Flash is cheaper: $1.50 against $4.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Sonnet 5.5 or Gemini 3.6 Flash?
- Gemini 3.6 Flash streams faster: 197 against 141 output tokens per second.
- Which has the larger context window?
- Gemini 3.6 Flash accepts more context: 1.0M against 1M tokens.