Model comparison
Claude Sonnet 5.5 vs GPT-5.2
Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 54.1 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude Sonnet 5.5 scores higher in 9 categories and GPT-5.2 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 5.5 leads 87.9 to 60.0.
- The biggest single-benchmark swing is ProofBench: 100% for Claude Sonnet 5.5 and 15% for GPT-5.2.
- Claude Sonnet 5.5 is cheaper at $2 / $10 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
- Claude Sonnet 5.5 accepts more context: 1M tokens versus 400K.
Side by side
| Claude Sonnet 5.5 | GPT-5.2 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 61.9 | 54.1 |
| Released | 2026-09-28 | 2025-12-11 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $2 | $1.75 |
| Output $ / M tokens | $10 | $14 |
| Results tracked | 32 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 67.3 (#6), GPT-5.2: 51.6 (#37)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| LMArena WebDev | 1774 | 1416 |
| LMArena Coding | 1513 | 1447 |
| ALE-Bench | 1,819 | 1,294 |
| SWE-bench Verified | — | 73.8% |
| FrontierCode | 52.1% | — |
| SWE-bench Verified (bash only) | — | 72.8% |
| CursorBench | 55.5% | — |
| SWE-bench Multilingual | — | 66.7% |
| FrontierSWE | 61.9% | — |
| SciCode | 61% | — |
| GSO | — | 27.4% |
| WeirdML | — | 72.2% |
| AlgoTune | — | 2.05 |
Agentic & Tool Use Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 45.0 (#16), GPT-5.2: 40.2 (#24)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| Terminal-Bench | — | 64.9% |
| APEX-Agents | 75.5% | — |
| 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% |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 75.3% |
| Vending-Bench 2 | — | 3,591 |
Reasoning Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 54.0 (#28), GPT-5.2: 50.2 (#35)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| NYT Connections (extended) | 80.5% | 83.6% |
| LMArena Hard Prompts | 1495 | 1445 |
| Mystery Game Puzzles | 65% | 23% |
| Epoch Capabilities Index | 165.03 | 153.45 |
| ARC-AGI-2 | — | 52.9% |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | — | 73.3% |
| ARC-AGI-1 | — | 86.2% |
| CritPt | 31.4% | — |
| Chess Puzzles | — | 49% |
| EnigmaEval | — | 10.4% |
| EBR-Bench | — | 23% |
| DTBench | — | 90.9% |
| LMCA | — | 43.9% |
| ForecastBench | — | 60.1 |
Math Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 87.9 (#6), GPT-5.2: 60.0 (#38)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 88.8% | 67.4% |
| FrontierMath Tier 4 | 80.5% | 31.7% |
| OTIS Mock AIME 2024-2025 | 100% | 96.1% |
| ProofBench | 100% | 15% |
| LMArena Math | 1510 | 1440 |
| MathArena Final-Answer Competitions | — | 72% |
| FrontierMath (Feb 2025 set) | — | 40.7% |
| FrontierMath Erdős | 2.9% | — |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 66.0 (#12), GPT-5.2: 59.3 (#32)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 95.6% | 91.4% |
| SimpleQA Verified | 46.5% | 37.1% |
| LMArena Expert | 1540 | 1445 |
| Humanity's Last Exam | — | 27.8% |
| Vectara Hallucination Rate | — | 8.4% |
Multimodal Too close to call
Claude Sonnet 5.5: 51.5 (#6), GPT-5.2: 51.3 (#7)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1289 | 1268 |
| Furniture Assembly | 75% | 38.3% |
| VPCT | — | 84% |
| LMArena Document | — | 1405 |
Multilingual Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 55.3 (#30), GPT-5.2: 53.4 (#67)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1452 | 1425 |
| LMArena Chinese | 1522 | 1460 |
| LMArena Russian | 1451 | 1440 |
| LMArena French | — | 1455 |
| LMArena German | — | 1448 |
| LMArena Japanese | — | 1420 |
| LMArena Korean | — | 1392 |
| LMArena Spanish | — | 1433 |
Instruction Following Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 78.3 (#11), GPT-5.2: 74.7 (#89)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1495 | 1417 |
Long Context Claude Sonnet 5.5 leads
Claude Sonnet 5.5: 45.9 (#28), GPT-5.2: 44.0 (#78)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Longer Query | 1498 | 1428 |
| CL-bench | — | 18.2% |
Writing & Preference Too close to call
Claude Sonnet 5.5: 66.0 (#40), GPT-5.2: 66.8 (#32)
| Benchmark | Claude Sonnet 5.5 | GPT-5.2 |
|---|---|---|
| LMArena Text | 1471 | 1439 |
| LMArena Creative Writing | 1465 | 1401 |
| LMArena Multi-Turn | 1474 | 1458 |
| EQ-Bench Creative Writing | — | 1703 |
Frequently asked questions
Is Claude Sonnet 5.5 better than GPT-5.2?
Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 54.1 on the Noometry Index.
Which is cheaper, Claude Sonnet 5.5 or GPT-5.2?
Claude Sonnet 5.5 is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; GPT-5.2 lists at $1.75 and $14.
Is Claude Sonnet 5.5 or GPT-5.2 better for coding?
Claude Sonnet 5.5 scores higher on coding benchmarks: 67.3 versus 51.6 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 5.5 does, with 1M tokens against 400K.
How many benchmarks do Claude Sonnet 5.5 and GPT-5.2 share?
25 benchmarks have published results for both models. Claude Sonnet 5.5 has 32 scored results on Noometry and GPT-5.2 has 67.