Model comparison
Claude Opus 4 vs GPT-4.1
Claude Opus 4 is the stronger model overall, scoring 43.1 to 35.9 on the Noometry Index. GPT-4.1 costs 8.6× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.
Last verified . 46 shared benchmarks.
Summary
- They share 46 benchmarks with published results for both. Claude Opus 4 scores higher in 7 categories and GPT-4.1 in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4 leads 42.0 to 22.3.
- The biggest single-benchmark swing is SimpleBench: 58.8% for Claude Opus 4 and 27% for GPT-4.1.
- GPT-4.1 is cheaper at $2 / $8 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- GPT-4.1 accepts more context: 1.05M tokens versus 200K.
Side by side
| Claude Opus 4 | GPT-4.1 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 43.1 | 35.9 |
| Released | 2025-05-22 | 2025-04-14 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 32K | 33K |
| Input $ / M tokens | $15 | $2 |
| Output $ / M tokens | $75 | $8 |
| Results tracked | 56 | 52 |
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-4.1: 34.4 (#238)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| SWE-bench Verified | 70.7% | 48.5% |
| SWE-bench Verified (bash only) | 67.6% | 39.6% |
| Aider Polyglot | 72% | 52.4% |
| WeirdML | 43.7% | 39% |
| LMArena Coding | 1442 | 1391 |
| GSO | 6.9% | — |
| CadEval | — | 42% |
| ALE-Bench | — | 558.1 |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Too close to call
Claude Opus 4: 34.8 (#42), GPT-4.1: 34.7 (#43)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 54% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
Reasoning Claude Opus 4 leads
Claude Opus 4: 27.3 (#121), GPT-4.1: 11.7 (#339)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| ARC-AGI-2 | 8.6% | 0.4% |
| SimpleBench | 58.8% | 27% |
| Kagi LLM Benchmark | 74.3% | 52.3% |
| ARC-AGI-1 | 35.7% | 5.5% |
| EnigmaEval | 5.6% | 2.2% |
| LMArena Hard Prompts | 1399 | 1384 |
| DTBench | 81.6% | 68.3% |
| LMCA | 37.4% | 25.6% |
| Epoch Capabilities Index | 142.67 | 136.78 |
| ForecastBench | 61.1 | 61.5 |
| CritPt | 0.3% | — |
| Chess Puzzles | — | 6% |
Math Claude Opus 4 leads
Claude Opus 4: 42.0 (#86), GPT-4.1: 22.3 (#280)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 38.3% |
| Omni-MATH | 61.6% | 47.1% |
| LMArena Math | 1390 | 1370 |
| MATH Level 5 | 85% | 83% |
| FrontierMath (Feb 2025 set) | 4.5% | 5.5% |
| FrontierMath Tier 4 (v1) | 4.2% | 0% |
| FrontierMath (Tiers 1-3) | — | 6% |
Knowledge Claude Opus 4 leads
Claude Opus 4: 44.0 (#88), GPT-4.1: 37.1 (#160)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 76.3% | 66.9% |
| Humanity's Last Exam | 10.7% | 5.4% |
| MMLU-Pro | 87.5% | 81.1% |
| Vectara Hallucination Rate | 12% | 5.6% |
| GPQA (HELM) | 70.8% | 65.9% |
| LMArena Expert | 1386 | 1364 |
| SimpleQA Verified | — | 31.1% |
| Confabulations | 15.9% | — |
Multimodal GPT-4.1 leads
Claude Opus 4: 31.5 (#106), GPT-4.1: 38.2 (#67)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| LMArena Vision | 1192 | 1211 |
| GeoBench | 49% | 72% |
| VPCT | 38% | — |
Multilingual Too close to call
Claude Opus 4: 48.8 (#138), GPT-4.1: 49.4 (#133)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1362 | 1370 |
| LMArena Chinese | 1386 | 1382 |
| LMArena French | 1372 | 1382 |
| LMArena German | 1391 | 1381 |
| LMArena Japanese | 1331 | 1319 |
| LMArena Korean | 1321 | 1339 |
| LMArena Russian | 1392 | 1377 |
| LMArena Spanish | 1389 | 1376 |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), GPT-4.1: 71.3 (#153)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| IFEval | 91.8% | 83.8% |
| LMArena Instruction Following | 1406 | 1367 |
Long Context Too close to call
Claude Opus 4: 39.6 (#172), GPT-4.1: 40.0 (#163)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| Fiction.LiveBench | 61.1% | 63.9% |
| LMArena Longer Query | 1422 | 1385 |
Writing & Preference Claude Opus 4 leads
Claude Opus 4: 61.2 (#89), GPT-4.1: 57.6 (#125)
| Benchmark | Claude Opus 4 | GPT-4.1 |
|---|---|---|
| LMArena Text | 1377 | 1383 |
| LMArena Creative Writing | 1387 | 1363 |
| EQ-Bench Creative Writing | 1580 | 1420 |
| WildBench | 85.2% | 85.4% |
| LMArena Multi-Turn | 1396 | 1398 |
| Short-Story Creative Writing | 83.6% | — |
Frequently asked questions
Is Claude Opus 4 better than GPT-4.1?
Claude Opus 4 is the stronger model overall, scoring 43.1 to 35.9 on the Noometry Index. GPT-4.1 costs 8.6× 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-4.1?
GPT-4.1 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or GPT-4.1 better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 200K.
How many benchmarks do Claude Opus 4 and GPT-4.1 share?
46 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and GPT-4.1 has 52.