Model comparison
Claude Opus 4.5 vs GPT-5.5
GPT-5.5 is the stronger model overall, scoring 63.4 to 50.5 on the Noometry Index.
Last verified . 52 shared benchmarks.
Summary
- They share 52 benchmarks with published results for both. Claude Opus 4.5 scores higher in 0 categories and GPT-5.5 in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.5 leads 81.7 to 38.6.
- The biggest single-benchmark swing is FrontierMath Tier 4: 4.9% for Claude Opus 4.5 and 72.5% for GPT-5.5.
- Claude Opus 4.5 is cheaper at $5 / $25 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 200K.
Side by side
| Claude Opus 4.5 | GPT-5.5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.5 | 63.4 |
| Released | 2025-11-01 | 2026-04-23 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 64K | 128K |
| Input $ / M tokens | $5 | $5 |
| Output $ / M tokens | $25 | $30 |
| Results tracked | 69 | 71 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
Claude Opus 4.5: 54.8 (#27), GPT-5.5: 58.2 (#17)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| SWE-bench Verified | 76.7% | 80.6% |
| LMArena WebDev | 1494 | 1513 |
| GSO | 26.5% | 40.2% |
| WeirdML | 63.7% | 84.9% |
| LMArena Coding | 1504 | 1494 |
| ALE-Bench | 1,025 | 1,943 |
| DeepSWE | — | 67% |
| FrontierCode | — | 43% |
| SWE-bench Verified (bash only) | 76.8% | — |
| SWE-bench Multilingual | 70.7% | — |
| SciCode | — | 56.1% |
| MirrorCode | — | 10% |
| AlgoTune | 1.77 | — |
Agentic & Tool Use GPT-5.5 leads
Claude Opus 4.5: 47.3 (#12), GPT-5.5: 50.7 (#6)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| Terminal-Bench | 63.1% | 84.7% |
| Remote Labor Index | 3.8% | 6.3% |
| τ²-bench Banking | 24.7% | 44.6% |
| DeepResearch Bench | 54.8% | 54% |
| LMArena Search | 1180 | 1242 |
| Vending-Bench 2 | 4,967 | 7,524 |
| APEX-Agents | — | 55.1% |
| Berkeley Function Calling Leaderboard | 77.5% | — |
| OSWorld 2.0 | — | 13% |
| GDPval | 45.5% | — |
| τ²-bench Airline | 84% | — |
| τ²-bench Retail | 79.6% | — |
| τ²-bench Telecom | 92.3% | — |
| Cybench | 82% | — |
| OSWorld | 66.3% | — |
| PostTrainBench | — | 27.2% |
| BALROG | 43.5% | — |
| ExploitBench | — | 47.4% |
| GBAEval | — | 53.2% |
| GDP.pdf | — | 26% |
| METR Time Horizons | 75% | — |
Reasoning GPT-5.5 leads
Claude Opus 4.5: 42.6 (#51), GPT-5.5: 72.8 (#11)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| ARC-AGI-2 | 37.6% | 85% |
| SimpleBench | 62% | 69% |
| Kagi LLM Benchmark | 80.2% | 88.8% |
| NYT Connections (extended) | 52.5% | 96.2% |
| ARC-AGI-1 | 80% | 95% |
| Chess Puzzles | 12% | 54% |
| EBR-Bench | 14.3% | 34.3% |
| LMArena Hard Prompts | 1476 | 1489 |
| Mystery Game Puzzles | 22% | 56% |
| DTBench | 89.9% | 96% |
| LMCA | 44.5% | 54.3% |
| Epoch Capabilities Index | 150.09 | 159.1 |
| ForecastBench | 60.7 | 60.6 |
| CritPt | — | 27.1% |
| EnigmaEval | 11.9% | — |
| Surface Evolver Bench | — | 88.1% |
| Bench to the Future 3 | — | 0.14 |
Math GPT-5.5 leads
Claude Opus 4.5: 38.6 (#132), GPT-5.5: 81.7 (#11)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 34.4% | 85.3% |
| FrontierMath Tier 4 | 4.9% | 72.5% |
| OTIS Mock AIME 2024-2025 | 86.1% | 100% |
| ProofBench | 36% | 50% |
| LMArena Math | 1463 | 1486 |
| FrontierMath (Feb 2025 set) | 20.7% | 51.7% |
| FrontierMath Tier 4 (v1) | 4.2% | 35.4% |
| MathArena Final-Answer Competitions | — | 94.3% |
| FrontierMath Erdős | — | 0% |
Knowledge GPT-5.5 leads
Claude Opus 4.5: 56.5 (#44), GPT-5.5: 64.4 (#17)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 86% | 94% |
| SimpleQA Verified | 45.7% | 63% |
| Vectara Hallucination Rate | 10.9% | 9.3% |
| LMArena Expert | 1487 | 1508 |
| Humanity's Last Exam | 25.2% | — |
Multimodal GPT-5.5 leads
Claude Opus 4.5: 31.4 (#107), GPT-5.5: 46.9 (#12)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| Furniture Assembly | 28.3% | 44.2% |
| LMArena Document | 1462 | 1486 |
| LMArena Vision | — | 1297 |
| GeoBench | 75% | — |
| VPCT | 40% | — |
| Blueprint-Bench 2 | — | 36.2% |
Multilingual GPT-5.5 leads
Claude Opus 4.5: 54.3 (#47), GPT-5.5: 56.4 (#20)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1438 | 1467 |
| LMArena Chinese | 1470 | 1533 |
| LMArena French | 1471 | 1486 |
| LMArena German | 1449 | 1480 |
| LMArena Japanese | 1416 | 1498 |
| LMArena Korean | 1424 | 1460 |
| LMArena Russian | 1447 | 1473 |
| LMArena Spanish | 1458 | 1468 |
Instruction Following Too close to call
Claude Opus 4.5: 77.5 (#19), GPT-5.5: 77.5 (#18)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1478 | 1479 |
Long Context GPT-5.5 leads
Claude Opus 4.5: 46.5 (#22), GPT-5.5: 48.3 (#12)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1480 | 1484 |
| CL-bench | 21.1% | — |
| CL-bench Life | — | 22.2% |
Writing & Preference GPT-5.5 leads
Claude Opus 4.5: 68.1 (#28), GPT-5.5: 72.7 (#13)
| Benchmark | Claude Opus 4.5 | GPT-5.5 |
|---|---|---|
| LMArena Text | 1451 | 1472 |
| LMArena Creative Writing | 1445 | 1455 |
| EQ-Bench Creative Writing | 1687 | 1844 |
| LMArena Multi-Turn | 1466 | 1476 |
| EQ-Bench 4 | — | 1315 |
Frequently asked questions
Is Claude Opus 4.5 better than GPT-5.5?
GPT-5.5 is the stronger model overall, scoring 63.4 to 50.5 on the Noometry Index.
Which is cheaper, Claude Opus 4.5 or GPT-5.5?
Claude Opus 4.5 is cheaper. It lists at $5 per million input tokens and $25 per million output tokens; GPT-5.5 lists at $5 and $30.
Is Claude Opus 4.5 or GPT-5.5 better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.2 versus 54.8 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 200K.
How many benchmarks do Claude Opus 4.5 and GPT-5.5 share?
52 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and GPT-5.5 has 71.