Model comparison
Claude Opus 4.8 vs GPT-4.5
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 37.2 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude Opus 4.8 scores higher in 10 categories and GPT-4.5 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.8 leads 64.7 to 13.9.
- The biggest single-benchmark swing is ARC-AGI-1: 92.5% for Claude Opus 4.8 and 10.3% for GPT-4.5.
Side by side
| Claude Opus 4.8 | GPT-4.5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 60.7 | 37.2 |
| Released | 2026-05-28 | 2025-02-27 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | — |
| Max output | 128K | — |
| Input $ / M tokens | $5 | — |
| Output $ / M tokens | $25 | — |
| Results tracked | 65 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.8 leads
Claude Opus 4.8: 59.9 (#12), GPT-4.5: 42.2 (#109)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| WeirdML | 82.9% | 39.4% |
| LMArena Coding | 1490 | 1396 |
| DeepSWE | 59% | — |
| FrontierCode | 46.5% | — |
| Aider Polyglot | — | 44.9% |
| LMArena WebDev | 1556 | — |
| SciCode | 53.5% | — |
| GSO | 47.1% | — |
| LiveBench Coding | — | 75.2% |
| ALE-Bench | 1,564 | — |
Agentic & Tool Use Claude Opus 4.8 leads
Claude Opus 4.8: 47.6 (#11), GPT-4.5: 27.9 (#97)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| APEX-Agents | 48.9% | — |
| OSWorld 2.0 | 20.6% | — |
| Remote Labor Index | 8.3% | — |
| τ²-bench Banking | 39.7% | — |
| Cybench | — | 17.5% |
| DeepResearch Bench | 50.2% | — |
| PostTrainBench | 33.8% | — |
| GBAEval | 70.9% | — |
| GDP.pdf | 24% | — |
| LMArena Search | 1204 | — |
| Vending-Bench 2 | 5,787 | — |
Reasoning Claude Opus 4.8 leads
Claude Opus 4.8: 64.7 (#16), GPT-4.5: 13.9 (#330)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| ARC-AGI-2 | 72.1% | 0.8% |
| SimpleBench | 64.8% | 34.5% |
| ARC-AGI-1 | 92.5% | 10.3% |
| EnigmaEval | 23.5% | 3.2% |
| LMArena Hard Prompts | 1482 | 1403 |
| Epoch Capabilities Index | 158.21 | 136.74 |
| ForecastBench | 59.9 | 61.7 |
| Kagi LLM Benchmark | 88.8% | — |
| NYT Connections (extended) | 91.1% | — |
| CritPt | 20.9% | — |
| Chess Puzzles | 34% | — |
| EBR-Bench | 28.6% | — |
| LiveBench Reasoning | — | 71.1% |
| Mystery Game Puzzles | 36% | — |
| DTBench | 94.9% | — |
| LiveBench Data Analysis | — | 64.3% |
| LMCA | 57.5% | — |
| Surface Evolver Bench | 87.5% | — |
| Bench to the Future 3 | 0.14 | — |
| LiveBench | — | 69% |
Math Claude Opus 4.8 leads
Claude Opus 4.8: 78.4 (#13), GPT-4.5: 32.6 (#211)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 37.8% |
| LMArena Math | 1487 | 1412 |
| FrontierMath (Tiers 1-3) | 80% | — |
| FrontierMath Tier 4 | 56.1% | — |
| MathArena Final-Answer Competitions | 91.8% | — |
| ProofBench | 69% | — |
| LiveBench Math | — | 69.3% |
| MATH Level 5 | — | 78.6% |
| FrontierMath (Feb 2025 set) | 47.2% | — |
| FrontierMath Tier 4 (v1) | 31.3% | — |
Knowledge Claude Opus 4.8 leads
Claude Opus 4.8: 61.3 (#29), GPT-4.5: 32.5 (#211)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| GPQA Diamond | 91% | 68.7% |
| LMArena Expert | 1502 | 1394 |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | 53% | — |
| Confabulations | — | 13.6% |
Multimodal Claude Opus 4.8 leads
Claude Opus 4.8: 42.9 (#26), GPT-4.5: 37.6 (#71)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| LMArena Vision | 1294 | 1195 |
| VPCT | — | 45% |
| Blueprint-Bench 2 | 14.5% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1475 | — |
Multilingual Claude Opus 4.8 leads
Claude Opus 4.8: 55.2 (#33), GPT-4.5: 52.5 (#83)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1450 | 1413 |
| LMArena Chinese | 1507 | 1421 |
| LMArena French | 1481 | 1418 |
| LMArena German | 1472 | 1457 |
| LMArena Japanese | 1440 | 1416 |
| LMArena Korean | 1432 | 1392 |
| LMArena Russian | 1474 | 1419 |
| LMArena Spanish | 1466 | — |
Instruction Following Claude Opus 4.8 leads
Claude Opus 4.8: 77.4 (#24), GPT-4.5: 72.6 (#134)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| LMArena Instruction Following | 1476 | 1404 |
| LiveBench Instruction Following | — | 72.3% |
Long Context Claude Opus 4.8 leads
Claude Opus 4.8: 45.4 (#35), GPT-4.5: 40.4 (#155)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| LMArena Longer Query | 1483 | 1406 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference Claude Opus 4.8 leads
Claude Opus 4.8: 72.0 (#16), GPT-4.5: 56.9 (#134)
| Benchmark | Claude Opus 4.8 | GPT-4.5 |
|---|---|---|
| LMArena Text | 1461 | 1417 |
| LMArena Creative Writing | 1454 | 1394 |
| EQ-Bench Creative Writing | 1840 | 1258 |
| LMArena Multi-Turn | 1476 | 1444 |
| Short-Story Creative Writing | — | 75.6% |
| EQ-Bench 4 | 1281 | — |
| LiveBench Language | — | 61.5% |
Frequently asked questions
Is Claude Opus 4.8 better than GPT-4.5?
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 37.2 on the Noometry Index.
Is Claude Opus 4.8 or GPT-4.5 better for coding?
Claude Opus 4.8 scores higher on coding benchmarks: 59.9 versus 42.2 in the Noometry coding category.
How many benchmarks do Claude Opus 4.8 and GPT-4.5 share?
27 benchmarks have published results for both models. Claude Opus 4.8 has 65 scored results on Noometry and GPT-4.5 has 42.