Model comparison
Claude Opus 5.5 vs Claude Sonnet 4.6
Claude Opus 5.5 is the stronger model overall, scoring 68.5 to 50.1 on the Noometry Index.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. Claude Opus 5.5 scores higher in 10 categories and Claude Sonnet 4.6 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 5.5 leads 92.4 to 51.9.
- The biggest single-benchmark swing is Mystery Game Puzzles: 71% for Claude Opus 5.5 and 16% for Claude Sonnet 4.6.
- Claude Sonnet 4.6 is cheaper at $3 / $15 per million input/output tokens, against $4 / $20 for Claude Opus 5.5.
Side by side
| Claude Opus 5.5 | Claude Sonnet 4.6 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 68.5 | 50.1 |
| Released | 2026-09-22 | 2026-02-17 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Max output | 128K | 128K |
| Input $ / M tokens | $4 | $3 |
| Output $ / M tokens | $20 | $15 |
| Results tracked | 35 | 57 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5.5 leads
Claude Opus 5.5: 71.4 (#1), Claude Sonnet 4.6: 46.0 (#62)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| FrontierCode | 54.4% | 24.3% |
| LMArena WebDev | 1820 | 1521 |
| SciCode | 66.9% | 46.8% |
| LMArena Coding | 1545 | 1504 |
| SWE-bench Verified | — | 75.2% |
| DeepSWE | — | 29.9% |
| CursorBench | 57.8% | — |
| FrontierSWE | 62.3% | — |
| WeirdML | — | 66.1% |
| MirrorCode | 77.4% | — |
| ALE-Bench | — | 1,327 |
Agentic & Tool Use Claude Opus 5.5 leads
Claude Opus 5.5: 45.3 (#13), Claude Sonnet 4.6: 39.1 (#26)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| APEX-Agents | 73.5% | 43% |
| GDP.pdf | 30.6% | 18% |
| Terminal-Bench | — | 53.4% |
| OSWorld 2.0 | — | 9.3% |
| DeepResearch Bench | — | 54.9% |
| OSWorld | — | 72.1% |
| ExploitBench | — | 23.6% |
| GBAEval | — | 48.8% |
| LMArena Search | — | 1221 |
| Vending-Bench 2 | — | 7,204 |
Reasoning Claude Opus 5.5 leads
Claude Opus 5.5: 79.5 (#2), Claude Sonnet 4.6: 46.1 (#40)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| ARC-AGI-2 | 92.5% | 60.4% |
| NYT Connections (extended) | 88.5% | 80.9% |
| ARC-AGI-1 | 97.5% | 86.5% |
| CritPt | 31.7% | 3.1% |
| LMArena Hard Prompts | 1545 | 1485 |
| Mystery Game Puzzles | 71% | 16% |
| DTBench | 98.9% | 89.9% |
| LMCA | 68.2% | 46.5% |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 76.3% |
| EBR-Bench | 71.4% | — |
| Epoch Capabilities Index | — | 152.25 |
| ForecastBench | — | 62 |
Math Claude Opus 5.5 leads
Claude Opus 5.5: 92.4 (#2), Claude Sonnet 4.6: 51.9 (#41)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 85.8% |
| ProofBench | 100% | 45% |
| FrontierMath (Tiers 1-3) | 91.2% | — |
| FrontierMath Tier 4 | 95% | — |
| LMArena Math | — | 1463 |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Tier 4 (v1) | — | 8.3% |
Knowledge Claude Opus 5.5 leads
Claude Opus 5.5: 66.6 (#9), Claude Sonnet 4.6: 51.6 (#60)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| GPQA Diamond | 90.6% | 87.4% |
| SimpleQA Verified | 72.2% | 35.5% |
| LMArena Expert | 1564 | 1500 |
| Vectara Hallucination Rate | — | 10.6% |
Multimodal Claude Opus 5.5 leads
Claude Opus 5.5: 56.5 (#2), Claude Sonnet 4.6: 38.0 (#61)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Vision | — | 1283 |
| Blueprint-Bench 2 | — | 6.7% |
| Furniture Assembly | 83.3% | — |
| LMArena Document | — | 1482 |
Multilingual Claude Opus 5.5 leads
Claude Opus 5.5: 59.8 (#1), Claude Sonnet 4.6: 54.6 (#39)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Non-English | 1513 | 1440 |
| LMArena Chinese | 1572 | 1491 |
| LMArena Russian | 1532 | 1440 |
| LMArena French | — | 1465 |
| LMArena German | — | 1428 |
| LMArena Japanese | — | 1419 |
| LMArena Korean | — | 1411 |
| LMArena Spanish | — | 1465 |
Instruction Following Claude Opus 5.5 leads
Claude Opus 5.5: 80.1 (#2), Claude Sonnet 4.6: 77.6 (#23)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Instruction Following | 1535 | 1476 |
Long Context Claude Opus 5.5 leads
Claude Opus 5.5: 47.3 (#18), Claude Sonnet 4.6: 45.3 (#40)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Longer Query | 1540 | 1479 |
Writing & Preference Claude Opus 5.5 leads
Claude Opus 5.5: 78.7 (#3), Claude Sonnet 4.6: 70.4 (#22)
| Benchmark | Claude Opus 5.5 | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Text | 1518 | 1458 |
| LMArena Creative Writing | 1538 | 1435 |
| EQ-Bench Creative Writing | 2050 | 1810 |
| LMArena Multi-Turn | 1520 | 1464 |
| EQ-Bench 4 | — | 1207 |
Frequently asked questions
Is Claude Opus 5.5 better than Claude Sonnet 4.6?
Claude Opus 5.5 is the stronger model overall, scoring 68.5 to 50.1 on the Noometry Index.
Which is cheaper, Claude Opus 5.5 or Claude Sonnet 4.6?
Claude Sonnet 4.6 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; Claude Opus 5.5 lists at $4 and $20.
Is Claude Opus 5.5 or Claude Sonnet 4.6 better for coding?
Claude Opus 5.5 scores higher on coding benchmarks: 71.4 versus 46.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Opus 5.5 and Claude Sonnet 4.6 share?
28 benchmarks have published results for both models. Claude Opus 5.5 has 35 scored results on Noometry and Claude Sonnet 4.6 has 57.