Model comparison
Claude Opus 4 vs GPT-4o
Claude Opus 4 is the stronger model overall, scoring 43.1 to 28.6 on the Noometry Index. GPT-4o costs 6.9× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.
Last verified . 51 shared benchmarks.
Summary
- They share 51 benchmarks with published results for both. Claude Opus 4 scores higher in 9 categories and GPT-4o in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4 leads 42.0 to 10.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 64.4% for Claude Opus 4 and 6.4% for GPT-4o.
- GPT-4o is cheaper at $2.50 / $10 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Claude Opus 4 accepts more context: 200K tokens versus 128K.
Side by side
| Claude Opus 4 | GPT-4o | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 43.1 | 28.6 |
| Released | 2025-05-22 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 128K |
| Max output | 32K | 16K |
| Input $ / M tokens | $15 | $2.50 |
| Output $ / M tokens | $75 | $10 |
| Results tracked | 56 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4 leads
Claude Opus 4: 47.2 (#62), GPT-4o: 24.8 (#328)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| SWE-bench Verified | 70.7% | 31% |
| SWE-bench Verified (bash only) | 67.6% | 21.6% |
| Aider Polyglot | 72% | 45.3% |
| GSO | 6.9% | 0% |
| WeirdML | 43.7% | 25.1% |
| LMArena Coding | 1442 | 1297 |
| BigCodeBench Instruct | — | 51.1% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| AlgoTune | 1.33 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use Claude Opus 4 leads
Claude Opus 4: 34.8 (#42), GPT-4o: 21.0 (#141)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| Cybench | 38% | 12.5% |
| LMArena Search | 1127 | 1006 |
| METR Time Horizons | 63.9% | 40.8% |
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| DeepResearch Bench | 46.8% | — |
| BALROG | — | 32.3% |
Reasoning Claude Opus 4 leads
Claude Opus 4: 27.3 (#121), GPT-4o: 9.4 (#343)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 8.6% | 0% |
| SimpleBench | 58.8% | 17.8% |
| ARC-AGI-1 | 35.7% | 4.5% |
| CritPt | 0.3% | 0% |
| EnigmaEval | 5.6% | 0.8% |
| LMArena Hard Prompts | 1399 | 1281 |
| DTBench | 81.6% | 64.5% |
| LMCA | 37.4% | 16.6% |
| Epoch Capabilities Index | 142.67 | 128.97 |
| ForecastBench | 61.1 | 57.7 |
| Kagi LLM Benchmark | 74.3% | — |
| Chess Puzzles | — | 13% |
| LiveBench Reasoning | — | 55.8% |
| LiveBench Data Analysis | — | 60.9% |
| LiveBench | — | 55.3% |
Math Claude Opus 4 leads
Claude Opus 4: 42.0 (#86), GPT-4o: 10.6 (#312)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 6.4% |
| Omni-MATH | 61.6% | 29.3% |
| LMArena Math | 1390 | 1285 |
| MATH Level 5 | 85% | 53.3% |
| FrontierMath (Feb 2025 set) | 4.5% | 0.3% |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| LiveBench Math | — | 49.5% |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Claude Opus 4 leads
Claude Opus 4: 44.0 (#88), GPT-4o: 28.8 (#242)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| GPQA Diamond | 76.3% | 49.2% |
| Humanity's Last Exam | 10.7% | 2.7% |
| MMLU-Pro | 87.5% | 71.3% |
| Confabulations | 15.9% | 15.3% |
| Vectara Hallucination Rate | 12% | 9.6% |
| GPQA (HELM) | 70.8% | 52% |
| LMArena Expert | 1386 | 1250 |
| SimpleQA Verified | — | 26% |
| MMLU | — | 88.1% |
Multimodal GPT-4o leads
Claude Opus 4: 31.5 (#106), GPT-4o: 34.5 (#91)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| LMArena Vision | 1192 | 1137 |
| GeoBench | 49% | 71% |
| VPCT | 38% | 40% |
| Video-MME | — | 71.9% |
| ScienceQA | — | 88.5% |
Multilingual Claude Opus 4 leads
Claude Opus 4: 48.8 (#138), GPT-4o: 43.2 (#186)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| LMArena Non-English | 1362 | 1283 |
| LMArena Chinese | 1386 | 1277 |
| LMArena French | 1372 | 1304 |
| LMArena German | 1391 | 1282 |
| LMArena Japanese | 1331 | 1257 |
| LMArena Korean | 1321 | 1234 |
| LMArena Russian | 1392 | 1286 |
| LMArena Spanish | 1389 | 1292 |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), GPT-4o: 66.6 (#207)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| IFEval | 91.8% | 81.7% |
| LMArena Instruction Following | 1406 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
Long Context Too close to call
Claude Opus 4: 39.6 (#172), GPT-4o: 39.4 (#179)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| Fiction.LiveBench | 61.1% | 66.7% |
| LMArena Longer Query | 1422 | 1289 |
Writing & Preference Claude Opus 4 leads
Claude Opus 4: 61.2 (#89), GPT-4o: 52.6 (#166)
| Benchmark | Claude Opus 4 | GPT-4o |
|---|---|---|
| LMArena Text | 1377 | 1300 |
| LMArena Creative Writing | 1387 | 1292 |
| Short-Story Creative Writing | 83.6% | 81.8% |
| WildBench | 85.2% | 82.8% |
| LMArena Multi-Turn | 1396 | 1302 |
| EQ-Bench Creative Writing | 1580 | — |
| LiveBench Language | — | 47.6% |
Frequently asked questions
Is Claude Opus 4 better than GPT-4o?
Claude Opus 4 is the stronger model overall, scoring 43.1 to 28.6 on the Noometry Index. GPT-4o costs 6.9× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 4 or GPT-4o?
GPT-4o is cheaper. It lists at $2.50 per million input tokens and $10 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or GPT-4o better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 24.8 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 4 does, with 200K tokens against 128K.
How many benchmarks do Claude Opus 4 and GPT-4o share?
51 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and GPT-4o has 72.