Claude Sonnet 5 vs Claude Sonnet 5.5
Claude Sonnet 5 Verdict
- Claude Sonnet 5.5 (max) leads on quality: 67.6 vs 60.5.
- Claude Sonnet 5 (high) is stronger in human preference, multimodal.
- Claude Sonnet 5.5 (max) is stronger in agents & tools, coding, composite, knowledge, long context, maths, reasoning.
- They cost about the same ($4.00 per 1M blended).
- Claude Sonnet 5.5 (max) streams 2.3× faster (141 vs 62 tokens per second).
| Metric | Claude Sonnet 5 (high) | Claude Sonnet 5.5 (max) |
|---|---|---|
| BenchLeader Index | 60.5 | 67.6 |
| Agents & tools score | 57.9 | 69.5 |
| Coding score | 61.1 | 69.8 |
| Composite score | 65.4 | 72.6 |
| Human preference score | 65.3 | – |
| Knowledge score | 60.0 | 65.4 |
| Long context score | 63.5 | 66.6 |
| Maths score | 65.8 | 72.9 |
| Multimodal score | 61.3 | – |
| Reasoning score | 66.3 | 81.1 |
| Blended price $/M | $4.00 | $4.00 |
| Output speed | 62 tok/s | 141 tok/s |
| Time to first answer | 12.3 s | 487.1 s |
| Context window | 1M | 1M |
| GPQA Diamond | – | 95.6% |
| FrontierMath Tiers 1–3 | – | 88.8% |
| FrontierMath Tier 4 | – | 80.5% |
| OTIS Mock AIME | – | 100.0% |
| SimpleQA Verified | – | 46.5% |
| Terminal-Bench | – | 61.8% |
| SciCode | 54.3% | 61.0% |
| WeirdML | 68.8% | – |
| APEX-Agents | – | 75.5% |
| ProofBench | – | 100.0% |
| LMArena Text | 1461 | – |
| LMArena Hard Prompts | 1490 | – |
| LMArena Coding | 1519 | – |
| LMArena WebDev | 1541 | – |
| LMArena Vision | 1274 | – |
| LMArena Agent | 4.9 | 12 |
| LiveBench | – | 75.7% |
| LiveBench Reasoning | – | 91.6% |
| LiveBench Coding | – | 91.4% |
| LiveBench Agentic Coding | – | 56.3% |
| LiveBench Mathematics | – | 96.1% |
| LiveBench Data Analysis | – | 59.5% |
| LiveBench Language | – | 78.0% |
| LiveBench Instruction Following | – | 56.8% |
| AA Intelligence Index v4.3.2 | 31.7 | 56 |
| AA-LCR | 76.7% | 82.7% |
| AA-Omniscience | -3.7 | 32.3 |
| Humanity's Last Exam (AA) | 35.7% | 55.0% |
| SciCode (AA) | 54.3% | 61.0% |
| CritPt | 15.1% | 31.4% |
| GDPval-AA v2.1 | 37.9% | 67.0% |
| Analyst Agent (AA) | – | 57.5% |
| LMArena Maths | 1473 | – |
| LMArena Creative Writing | 1435 | – |
| LMArena Instruction Following | 1466 | – |
| LMArena Multi-turn | 1473 | – |
| LMArena Longer Queries | 1481 | – |
| LMArena Document | 1476 | – |
| Mystery Game Puzzles | – | 65.0% |
| DeepSWE v1.1 | 48.2% | – |
| LMCA | 50.0% | – |
| DTBench | 84.5% | – |
| CursorBench | 30.8% | 55.5% |
| ALE-Bench | 1463.1 | – |
| FrontierSWE | – | 61.9% |
| Terminal-Bench 4.0 (AA) | 5.0% | 63.6% |
| AutomationBench | – | 71.8% |
| GDP.pdf | – | 25.8% |
| MLCR | – | 75.0% |
| Harvey LAB | – | 2.8% |
| AA-Omniscience: accuracy | 37.4% | 54.0% |
| AA-Omniscience: non-hallucination | 34.3% | 53.0% |
| 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 vs Claude Sonnet 5.5: questions
- Is Claude Sonnet 5 better than Claude Sonnet 5.5?
- Claude Sonnet 5.5 (max) leads on quality: 67.6 vs 60.5. 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 better than Claude Sonnet 5.5 for coding?
- Claude Sonnet 5.5 scores higher in coding (70 vs 61 on the category index, where 50 is average).
- Is Claude Sonnet 5 better than Claude Sonnet 5.5 for agentic tasks?
- Claude Sonnet 5.5 scores higher in agentic tasks (70 vs 58 on the category index, where 50 is average).
- Which is cheaper, Claude Sonnet 5 or Claude Sonnet 5.5?
- Claude Sonnet 5 is cheaper: $4.00 against $4.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Sonnet 5 or Claude Sonnet 5.5?
- Claude Sonnet 5.5 streams faster: 141 against 62 output tokens per second.
- Which has the larger context window?
- Both accept 1M tokens of context.