Model comparison
Claude Opus 5.5 vs GPT-5.2
Claude Opus 5.5 is the stronger model overall, scoring 68.6 to 54.1 on the Noometry Index. GPT-5.2 costs 1.7× less per token, which makes it the better buy when Claude Opus 5.5's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. Claude Opus 5.5 scores higher in 10 categories and GPT-5.2 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 5.5 leads 91.8 to 60.0.
- The biggest single-benchmark swing is ProofBench: 100% for Claude Opus 5.5 and 15% for GPT-5.2.
- GPT-5.2 is cheaper at $1.75 / $14 per million input/output tokens, against $4 / $20 for Claude Opus 5.5.
- Claude Opus 5.5 accepts more context: 1M tokens versus 400K.
Side by side
| Claude Opus 5.5 | GPT-5.2 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 68.6 | 54.1 |
| Released | 2026-09-22 | 2025-12-11 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $4 | $1.75 |
| Output $ / M tokens | $20 | $14 |
| Results tracked | 44 | 67 |
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-5.2: 51.6 (#37)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| LMArena WebDev | 1813 | 1416 |
| LMArena Coding | 1547 | 1447 |
| ALE-Bench | 2,147 | 1,294 |
| SWE-bench Verified | — | 73.8% |
| FrontierCode | 54.6% | — |
| SWE-bench Verified (bash only) | — | 72.8% |
| CursorBench | 57.8% | — |
| SWE-bench Multilingual | — | 66.7% |
| FrontierSWE | 62.3% | — |
| SciCode | 66.9% | — |
| GSO | — | 27.4% |
| WeirdML | — | 72.2% |
| MirrorCode | 77.4% | — |
| AlgoTune | — | 2.05 |
Agentic & Tool Use Claude Opus 5.5 leads
Claude Opus 5.5: 45.3 (#15), GPT-5.2: 40.2 (#24)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| Vending-Bench 2 | 9,235 | 3,591 |
| Terminal-Bench | — | 64.9% |
| APEX-Agents | 73.5% | — |
| Berkeley Function Calling Leaderboard | — | 55.9% |
| GDPval | — | 49.7% |
| Remote Labor Index | — | 2.5% |
| τ²-bench Airline | — | 83% |
| τ²-bench Banking | — | 32.2% |
| τ²-bench Retail | — | 81.6% |
| τ²-bench Telecom | — | 89.7% |
| DeepResearch Bench | — | 41.1% |
| GDP.pdf | 30.6% | — |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 75.3% |
Reasoning Claude Opus 5.5 leads
Claude Opus 5.5: 80.2 (#3), GPT-5.2: 50.2 (#35)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| ARC-AGI-2 | 93.3% | 52.9% |
| NYT Connections (extended) | 88.5% | 83.6% |
| ARC-AGI-1 | 98.5% | 86.2% |
| EBR-Bench | 71.4% | 23% |
| LMArena Hard Prompts | 1535 | 1445 |
| Mystery Game Puzzles | 71% | 23% |
| DTBench | 98.9% | 90.9% |
| LMCA | 68.2% | 43.9% |
| Epoch Capabilities Index | 167.33 | 153.45 |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | — | 73.3% |
| CritPt | 31.7% | — |
| Chess Puzzles | — | 49% |
| EnigmaEval | — | 10.4% |
| ForecastBench | — | 60.1 |
Math Claude Opus 5.5 leads
Claude Opus 5.5: 91.8 (#3), GPT-5.2: 60.0 (#38)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 91.2% | 67.4% |
| FrontierMath Tier 4 | 95% | 31.7% |
| OTIS Mock AIME 2024-2025 | 100% | 96.1% |
| ProofBench | 100% | 15% |
| LMArena Math | 1506 | 1440 |
| MathArena Final-Answer Competitions | — | 72% |
| FrontierMath (Feb 2025 set) | — | 40.7% |
| FrontierMath Erdős | 2.9% | — |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge Claude Opus 5.5 leads
Claude Opus 5.5: 66.4 (#10), GPT-5.2: 59.3 (#32)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 90.6% | 91.4% |
| SimpleQA Verified | 72.2% | 37.1% |
| LMArena Expert | 1547 | 1445 |
| Humanity's Last Exam | — | 27.8% |
| Vectara Hallucination Rate | — | 8.4% |
Multimodal Claude Opus 5.5 leads
Claude Opus 5.5: 57.8 (#1), GPT-5.2: 51.3 (#7)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1321 | 1268 |
| Furniture Assembly | 83.3% | 38.3% |
| VPCT | — | 84% |
| Blueprint-Bench 2 | 51.2% | — |
| LMArena Document | — | 1405 |
Multilingual Claude Opus 5.5 leads
Claude Opus 5.5: 59.1 (#2), GPT-5.2: 53.4 (#67)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1507 | 1425 |
| LMArena Chinese | 1588 | 1460 |
| LMArena French | 1514 | 1455 |
| LMArena Russian | 1520 | 1440 |
| LMArena Spanish | 1507 | 1433 |
| LMArena German | — | 1448 |
| LMArena Japanese | — | 1420 |
| LMArena Korean | — | 1392 |
Instruction Following Claude Opus 5.5 leads
Claude Opus 5.5: 80.0 (#3), GPT-5.2: 74.7 (#89)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1537 | 1417 |
Long Context Claude Opus 5.5 leads
Claude Opus 5.5: 47.1 (#19), GPT-5.2: 44.0 (#78)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Longer Query | 1532 | 1428 |
| CL-bench | — | 18.2% |
Writing & Preference Claude Opus 5.5 leads
Claude Opus 5.5: 78.2 (#3), GPT-5.2: 66.8 (#32)
| Benchmark | Claude Opus 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Text | 1515 | 1439 |
| LMArena Creative Writing | 1533 | 1401 |
| EQ-Bench Creative Writing | 2050 | 1703 |
| LMArena Multi-Turn | 1499 | 1458 |
Frequently asked questions
Is Claude Opus 5.5 better than GPT-5.2?
Claude Opus 5.5 is the stronger model overall, scoring 68.6 to 54.1 on the Noometry Index. GPT-5.2 costs 1.7× less per token, which makes it the better buy when Claude Opus 5.5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5.5 or GPT-5.2?
GPT-5.2 is cheaper. It lists at $1.75 per million input tokens and $14 per million output tokens; Claude Opus 5.5 lists at $4 and $20.
Is Claude Opus 5.5 or GPT-5.2 better for coding?
Claude Opus 5.5 scores higher on coding benchmarks: 71.9 versus 51.6 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 5.5 does, with 1M tokens against 400K.
How many benchmarks do Claude Opus 5.5 and GPT-5.2 share?
34 benchmarks have published results for both models. Claude Opus 5.5 has 44 scored results on Noometry and GPT-5.2 has 67.