Model comparison
Claude Opus 5 vs GPT-5.1
Claude Opus 5 is the stronger model overall, scoring 67.7 to 48.8 on the Noometry Index. GPT-5.1 costs 2.9× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. Claude Opus 5 scores higher in 8 categories and GPT-5.1 in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5 leads 77.1 to 39.9.
- The biggest single-benchmark swing is ARC-AGI-2: 90.4% for Claude Opus 5 and 17.6% for GPT-5.1.
- GPT-5.1 is cheaper at $1.25 / $10 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Claude Opus 5 accepts more context: 1M tokens versus 400K.
Side by side
| Claude Opus 5 | GPT-5.1 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 67.7 | 48.8 |
| Released | 2026-07-24 | 2025-11-13 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $5 | $1.25 |
| Output $ / M tokens | $25 | $10 |
| Results tracked | 57 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude Opus 5: 67.0 (#5), GPT-5.1: 46.9 (#59)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| LMArena WebDev | 1694 | 1395 |
| SciCode | 56.4% | 43.3% |
| WeirdML | 91.8% | 60.8% |
| LMArena Coding | 1533 | 1454 |
| ALE-Bench | 2,165 | 1,192 |
| SWE-bench Verified | — | 68% |
| DeepSWE | 73.6% | — |
| FrontierCode | 53.4% | — |
| SWE-bench Verified (bash only) | — | 66% |
| CursorBench | 46.6% | — |
| FrontierSWE | 52% | — |
| GSO | — | 13.7% |
| LiveBench Coding | — | 72.5% |
Agentic & Tool Use Claude Opus 5 leads
Claude Opus 5: 55.5 (#1), GPT-5.1: 32.5 (#55)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| Vending-Bench 2 | 11,182 | 1,473 |
| Terminal-Bench | — | 47.6% |
| APEX-Agents | 65.8% | — |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| DeepResearch Bench | — | 42.8% |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| GDP.pdf | 24% | — |
| LMArena Search | — | 1199 |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.1 (#3), GPT-5.1: 39.9 (#50)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 90.4% | 17.6% |
| SimpleBench | 80.6% | 53.2% |
| ARC-AGI-1 | 97.5% | 72.8% |
| CritPt | 29.1% | 4.9% |
| Chess Puzzles | 42% | 32% |
| LMArena Hard Prompts | 1525 | 1457 |
| Mystery Game Puzzles | 59% | 19% |
| DTBench | 97.9% | 90.1% |
| LMCA | 64.5% | 43.9% |
| Epoch Capabilities Index | 162.67 | 149.66 |
| NYT Connections (extended) | 94.3% | — |
| EnigmaEval | — | 11.2% |
| EBR-Bench | 45.7% | — |
| LiveBench Reasoning | — | 95.8% |
| LiveBench Data Analysis | — | 72.1% |
| Bench to the Future 3 | 0.12 | — |
| ForecastBench | — | 58.1 |
| LiveBench | — | 78.8% |
Math Claude Opus 5 leads
Claude Opus 5: 86.1 (#6), GPT-5.1: 50.3 (#45)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 88.6% |
| LMArena Math | 1533 | 1446 |
| FrontierMath (Tiers 1-3) | 85.6% | — |
| FrontierMath Tier 4 | 73.2% | — |
| ProofBench | 99% | — |
| Omni-MATH | — | 46.4% |
| LiveBench Math | — | 94.5% |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge Claude Opus 5 leads
Claude Opus 5: 66.8 (#8), GPT-5.1: 50.5 (#65)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 93.9% | 87.6% |
| SimpleQA Verified | 59.9% | 48% |
| LMArena Expert | 1558 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| MMLU-Pro | — | 57.9% |
| Vectara Hallucination Rate | — | 10.9% |
| GPQA (HELM) | — | 44.2% |
Multimodal Claude Opus 5 leads
Claude Opus 5: 51.1 (#6), GPT-5.1: 44.0 (#17)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1321 | 1250 |
| LMArena Document | 1516 | 1403 |
| VPCT | — | 58.7% |
| Blueprint-Bench 2 | 30.4% | — |
| Furniture Assembly | 60.8% | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#3), GPT-5.1: 54.0 (#51)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1500 | 1431 |
| LMArena Chinese | 1575 | 1495 |
| LMArena French | 1522 | 1451 |
| LMArena German | 1522 | 1439 |
| LMArena Japanese | 1510 | 1453 |
| LMArena Korean | 1526 | 1401 |
| LMArena Russian | 1507 | 1435 |
| LMArena Spanish | 1519 | 1430 |
Instruction Following GPT-5.1 leads
Claude Opus 5: 79.3 (#5), GPT-5.1: 84.0 (#1)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| LMArena Instruction Following | 1516 | 1442 |
| LiveBench Instruction Following | — | 93.3% |
| IFEval | — | 93.5% |
Long Context GPT-5.1 leads
Claude Opus 5: 46.5 (#20), GPT-5.1: 47.6 (#14)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1516 | 1448 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.3 (#2), GPT-5.1: 64.7 (#53)
| Benchmark | Claude Opus 5 | GPT-5.1 |
|---|---|---|
| LMArena Text | 1506 | 1443 |
| LMArena Creative Writing | 1491 | 1427 |
| LMArena Multi-Turn | 1496 | 1449 |
| EQ-Bench Creative Writing | 2133 | — |
| WildBench | — | 86.3% |
| EQ-Bench 4 | 1385 | — |
| LiveBench Language | — | 80.2% |
Frequently asked questions
Is Claude Opus 5 better than GPT-5.1?
Claude Opus 5 is the stronger model overall, scoring 67.7 to 48.8 on the Noometry Index. GPT-5.1 costs 2.9× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5 or GPT-5.1?
GPT-5.1 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or GPT-5.1 better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.0 versus 46.9 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 5 does, with 1M tokens against 400K.
How many benchmarks do Claude Opus 5 and GPT-5.1 share?
36 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and GPT-5.1 has 63.