Model comparison
Claude Opus 4.5 vs Claude Sonnet 5.5
Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 50.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. Claude Opus 4.5 scores higher in 3 categories and Claude Sonnet 5.5 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 5.5 leads 87.9 to 38.6.
- The biggest single-benchmark swing is FrontierMath Tier 4: 4.9% for Claude Opus 4.5 and 80.5% for Claude Sonnet 5.5.
- Claude Sonnet 5.5 is cheaper at $2 / $10 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
- Claude Sonnet 5.5 accepts more context: 1M tokens versus 200K.
Side by side
| Claude Opus 4.5 | Claude Sonnet 5.5 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 50.5 | 61.9 |
| Released | 2025-11-01 | 2026-09-28 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1M |
| Max output | 64K | 128K |
| Input $ / M tokens | $5 | $2 |
| Output $ / M tokens | $25 | $10 |
| Results tracked | 69 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5.5 leads
Claude Opus 4.5: 54.8 (#27), Claude Sonnet 5.5: 67.3 (#6)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| LMArena WebDev | 1494 | 1774 |
| LMArena Coding | 1504 | 1513 |
| ALE-Bench | 1,025 | 1,819 |
| SWE-bench Verified | 76.7% | — |
| FrontierCode | — | 52.1% |
| SWE-bench Verified (bash only) | 76.8% | — |
| CursorBench | — | 55.5% |
| SWE-bench Multilingual | 70.7% | — |
| FrontierSWE | — | 61.9% |
| SciCode | — | 61% |
| GSO | 26.5% | — |
| WeirdML | 63.7% | — |
| AlgoTune | 1.77 | — |
Agentic & Tool Use Claude Opus 4.5 leads
Claude Opus 4.5: 47.3 (#12), Claude Sonnet 5.5: 45.0 (#16)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| Terminal-Bench | 63.1% | — |
| APEX-Agents | — | 75.5% |
| Berkeley Function Calling Leaderboard | 77.5% | — |
| GDPval | 45.5% | — |
| Remote Labor Index | 3.8% | — |
| τ²-bench Airline | 84% | — |
| τ²-bench Banking | 24.7% | — |
| τ²-bench Retail | 79.6% | — |
| τ²-bench Telecom | 92.3% | — |
| Cybench | 82% | — |
| DeepResearch Bench | 54.8% | — |
| OSWorld | 66.3% | — |
| BALROG | 43.5% | — |
| LMArena Search | 1180 | — |
| METR Time Horizons | 75% | — |
| Vending-Bench 2 | 4,967 | — |
Reasoning Claude Sonnet 5.5 leads
Claude Opus 4.5: 42.6 (#51), Claude Sonnet 5.5: 54.0 (#28)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| NYT Connections (extended) | 52.5% | 80.5% |
| LMArena Hard Prompts | 1476 | 1495 |
| Mystery Game Puzzles | 22% | 65% |
| Epoch Capabilities Index | 150.09 | 165.03 |
| ARC-AGI-2 | 37.6% | — |
| SimpleBench | 62% | — |
| Kagi LLM Benchmark | 80.2% | — |
| ARC-AGI-1 | 80% | — |
| CritPt | — | 31.4% |
| Chess Puzzles | 12% | — |
| EnigmaEval | 11.9% | — |
| EBR-Bench | 14.3% | — |
| DTBench | 89.9% | — |
| LMCA | 44.5% | — |
| ForecastBench | 60.7 | — |
Math Claude Sonnet 5.5 leads
Claude Opus 4.5: 38.6 (#132), Claude Sonnet 5.5: 87.9 (#6)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 34.4% | 88.8% |
| FrontierMath Tier 4 | 4.9% | 80.5% |
| OTIS Mock AIME 2024-2025 | 86.1% | 100% |
| ProofBench | 36% | 100% |
| LMArena Math | 1463 | 1510 |
| FrontierMath (Feb 2025 set) | 20.7% | — |
| FrontierMath Erdős | — | 2.9% |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Claude Sonnet 5.5 leads
Claude Opus 4.5: 56.5 (#44), Claude Sonnet 5.5: 66.0 (#12)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| GPQA Diamond | 86% | 95.6% |
| SimpleQA Verified | 45.7% | 46.5% |
| LMArena Expert | 1487 | 1540 |
| Humanity's Last Exam | 25.2% | — |
| Vectara Hallucination Rate | 10.9% | — |
Multimodal Claude Sonnet 5.5 leads
Claude Opus 4.5: 31.4 (#107), Claude Sonnet 5.5: 51.5 (#6)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| Furniture Assembly | 28.3% | 75% |
| LMArena Vision | — | 1289 |
| GeoBench | 75% | — |
| VPCT | 40% | — |
| LMArena Document | 1462 | — |
Multilingual Claude Sonnet 5.5 leads
Claude Opus 4.5: 54.3 (#47), Claude Sonnet 5.5: 55.3 (#30)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| LMArena Non-English | 1438 | 1452 |
| LMArena Chinese | 1470 | 1522 |
| LMArena Russian | 1447 | 1451 |
| LMArena French | 1471 | — |
| LMArena German | 1449 | — |
| LMArena Japanese | 1416 | — |
| LMArena Korean | 1424 | — |
| LMArena Spanish | 1458 | — |
Instruction Following Too close to call
Claude Opus 4.5: 77.5 (#19), Claude Sonnet 5.5: 78.3 (#11)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| LMArena Instruction Following | 1478 | 1495 |
Long Context Too close to call
Claude Opus 4.5: 46.5 (#22), Claude Sonnet 5.5: 45.9 (#28)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| LMArena Longer Query | 1480 | 1498 |
| CL-bench | 21.1% | — |
Writing & Preference Claude Opus 4.5 leads
Claude Opus 4.5: 68.1 (#28), Claude Sonnet 5.5: 66.0 (#40)
| Benchmark | Claude Opus 4.5 | Claude Sonnet 5.5 |
|---|---|---|
| LMArena Text | 1451 | 1471 |
| LMArena Creative Writing | 1445 | 1465 |
| LMArena Multi-Turn | 1466 | 1474 |
| EQ-Bench Creative Writing | 1687 | — |
Frequently asked questions
Is Claude Opus 4.5 better than Claude Sonnet 5.5?
Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 50.5 on the Noometry Index.
Which is cheaper, Claude Opus 4.5 or Claude Sonnet 5.5?
Claude Sonnet 5.5 is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; Claude Opus 4.5 lists at $5 and $25.
Is Claude Opus 4.5 or Claude Sonnet 5.5 better for coding?
Claude Sonnet 5.5 scores higher on coding benchmarks: 67.3 versus 54.8 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 5.5 does, with 1M tokens against 200K.
How many benchmarks do Claude Opus 4.5 and Claude Sonnet 5.5 share?
24 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and Claude Sonnet 5.5 has 32.