Model comparison
Claude Opus 5.5 vs GPT-5.5
Claude Opus 5.5 is the stronger model overall, scoring 68.6 to 63.4 on the Noometry Index.
Last verified . 42 shared benchmarks.
Summary
- They share 42 benchmarks with published results for both. Claude Opus 5.5 scores higher in 8 categories and GPT-5.5 in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude Opus 5.5 leads 71.9 to 58.2.
- The biggest single-benchmark swing is MirrorCode: 77.4% for Claude Opus 5.5 and 10% for GPT-5.5.
- Claude Opus 5.5 is cheaper at $4 / $20 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Opus 5.5 | GPT-5.5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 68.6 | 63.4 |
| Released | 2026-09-22 | 2026-04-23 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $4 | $5 |
| Output $ / M tokens | $20 | $30 |
| Results tracked | 44 | 71 |
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: 71.9 (#3), GPT-5.5: 58.2 (#17)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| FrontierCode | 54.6% | 43% |
| LMArena WebDev | 1813 | 1513 |
| SciCode | 66.9% | 56.1% |
| LMArena Coding | 1547 | 1494 |
| MirrorCode | 77.4% | 10% |
| ALE-Bench | 2,147 | 1,943 |
| SWE-bench Verified | — | 80.6% |
| DeepSWE | — | 67% |
| CursorBench | 57.8% | — |
| FrontierSWE | 62.3% | — |
| GSO | — | 40.2% |
| WeirdML | — | 84.9% |
Agentic & Tool Use GPT-5.5 leads
Claude Opus 5.5: 45.3 (#15), GPT-5.5: 50.7 (#6)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| APEX-Agents | 73.5% | 55.1% |
| GDP.pdf | 30.6% | 26% |
| Vending-Bench 2 | 9,235 | 7,524 |
| Terminal-Bench | — | 84.7% |
| OSWorld 2.0 | — | 13% |
| Remote Labor Index | — | 6.3% |
| τ²-bench Banking | — | 44.6% |
| DeepResearch Bench | — | 54% |
| PostTrainBench | — | 27.2% |
| ExploitBench | — | 47.4% |
| GBAEval | — | 53.2% |
| LMArena Search | — | 1242 |
Reasoning Claude Opus 5.5 leads
Claude Opus 5.5: 80.2 (#3), GPT-5.5: 72.8 (#11)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| ARC-AGI-2 | 93.3% | 85% |
| NYT Connections (extended) | 88.5% | 96.2% |
| ARC-AGI-1 | 98.5% | 95% |
| CritPt | 31.7% | 27.1% |
| EBR-Bench | 71.4% | 34.3% |
| LMArena Hard Prompts | 1535 | 1489 |
| Mystery Game Puzzles | 71% | 56% |
| DTBench | 98.9% | 96% |
| LMCA | 68.2% | 54.3% |
| Epoch Capabilities Index | 167.33 | 159.1 |
| SimpleBench | — | 69% |
| Kagi LLM Benchmark | — | 88.8% |
| Chess Puzzles | — | 54% |
| Surface Evolver Bench | — | 88.1% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | — | 60.6 |
Math Claude Opus 5.5 leads
Claude Opus 5.5: 91.8 (#3), GPT-5.5: 81.7 (#11)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 91.2% | 85.3% |
| FrontierMath Tier 4 | 95% | 72.5% |
| OTIS Mock AIME 2024-2025 | 100% | 100% |
| ProofBench | 100% | 50% |
| LMArena Math | 1506 | 1486 |
| FrontierMath Erdős | 2.9% | 0% |
| MathArena Final-Answer Competitions | — | 94.3% |
| FrontierMath (Feb 2025 set) | — | 51.7% |
| FrontierMath Tier 4 (v1) | — | 35.4% |
Knowledge Claude Opus 5.5 leads
Claude Opus 5.5: 66.4 (#10), GPT-5.5: 64.4 (#17)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 90.6% | 94% |
| SimpleQA Verified | 72.2% | 63% |
| LMArena Expert | 1547 | 1508 |
| Vectara Hallucination Rate | — | 9.3% |
Multimodal Claude Opus 5.5 leads
Claude Opus 5.5: 57.8 (#1), GPT-5.5: 46.9 (#12)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1321 | 1297 |
| Blueprint-Bench 2 | 51.2% | 36.2% |
| Furniture Assembly | 83.3% | 44.2% |
| LMArena Document | — | 1486 |
Multilingual Claude Opus 5.5 leads
Claude Opus 5.5: 59.1 (#2), GPT-5.5: 56.4 (#20)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1507 | 1467 |
| LMArena Chinese | 1588 | 1533 |
| LMArena French | 1514 | 1486 |
| LMArena Russian | 1520 | 1473 |
| LMArena Spanish | 1507 | 1468 |
| LMArena German | — | 1480 |
| LMArena Japanese | — | 1498 |
| LMArena Korean | — | 1460 |
Instruction Following Claude Opus 5.5 leads
Claude Opus 5.5: 80.0 (#3), GPT-5.5: 77.5 (#18)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1537 | 1479 |
Long Context GPT-5.5 leads
Claude Opus 5.5: 47.1 (#19), GPT-5.5: 48.3 (#12)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1532 | 1484 |
| CL-bench Life | — | 22.2% |
Writing & Preference Claude Opus 5.5 leads
Claude Opus 5.5: 78.2 (#3), GPT-5.5: 72.7 (#13)
| Benchmark | Claude Opus 5.5 | GPT-5.5 |
|---|---|---|
| LMArena Text | 1515 | 1472 |
| LMArena Creative Writing | 1533 | 1455 |
| EQ-Bench Creative Writing | 2050 | 1844 |
| LMArena Multi-Turn | 1499 | 1476 |
| EQ-Bench 4 | — | 1315 |
Frequently asked questions
Is Claude Opus 5.5 better than GPT-5.5?
Claude Opus 5.5 is the stronger model overall, scoring 68.6 to 63.4 on the Noometry Index.
Which is cheaper, Claude Opus 5.5 or GPT-5.5?
Claude Opus 5.5 is cheaper. It lists at $4 per million input tokens and $20 per million output tokens; GPT-5.5 lists at $5 and $30.
Is Claude Opus 5.5 or GPT-5.5 better for coding?
Claude Opus 5.5 scores higher on coding benchmarks: 71.9 versus 58.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 1M.
How many benchmarks do Claude Opus 5.5 and GPT-5.5 share?
42 benchmarks have published results for both models. Claude Opus 5.5 has 44 scored results on Noometry and GPT-5.5 has 71.