Model comparison
Claude Opus 4 vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 43.1 on the Noometry Index.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. Claude Opus 4 scores higher in 3 categories and GPT-5.4 mini in 7 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where GPT-5.4 mini leads 39.7 to 31.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 74.3% for Claude Opus 4 and 37.9% for GPT-5.4 mini.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- GPT-5.4 mini accepts more context: 400K tokens versus 200K.
Side by side
| Claude Opus 4 | GPT-5.4 mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 43.1 | 45.0 |
| Released | 2025-05-22 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 400K |
| Max output | 32K | 128K |
| Input $ / M tokens | $15 | $0.75 |
| Output $ / M tokens | $75 | $4.50 |
| Results tracked | 56 | 46 |
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-5.4 mini: 45.2 (#72)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| WeirdML | 43.7% | 60.3% |
| LMArena Coding | 1442 | 1438 |
| SWE-bench Verified | 70.7% | — |
| FrontierCode | — | 27% |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1397 |
| SciCode | — | 49.9% |
| GSO | 6.9% | — |
| ALE-Bench | — | 1,189 |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Claude Opus 4 leads
Claude Opus 4: 34.8 (#42), GPT-5.4 mini: 29.9 (#81)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | 46.8% | 36.3% |
| Cybench | 38% | — |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
Reasoning GPT-5.4 mini leads
Claude Opus 4: 27.3 (#121), GPT-5.4 mini: 30.4 (#85)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| ARC-AGI-2 | 8.6% | 18.9% |
| Kagi LLM Benchmark | 74.3% | 37.9% |
| ARC-AGI-1 | 35.7% | 63.7% |
| CritPt | 0.3% | 10% |
| LMArena Hard Prompts | 1399 | 1424 |
| DTBench | 81.6% | 80% |
| LMCA | 37.4% | 40.8% |
| Epoch Capabilities Index | 142.67 | 148.84 |
| ForecastBench | 61.1 | 57 |
| SimpleBench | 58.8% | — |
| NYT Connections (extended) | — | 61.8% |
| Chess Puzzles | — | 24% |
| EnigmaEval | 5.6% | — |
| Thematic Generalization | — | 61.7% |
| Mystery Game Puzzles | — | 11% |
Math GPT-5.4 mini leads
Claude Opus 4: 42.0 (#86), GPT-5.4 mini: 45.5 (#75)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 88.9% |
| LMArena Math | 1390 | 1419 |
| FrontierMath (Feb 2025 set) | 4.5% | 28.3% |
| FrontierMath Tier 4 (v1) | 4.2% | 2.1% |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| ProofBench | — | 21% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
Knowledge GPT-5.4 mini leads
Claude Opus 4: 44.0 (#88), GPT-5.4 mini: 51.5 (#67)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 76.3% | 86.9% |
| Vectara Hallucination Rate | 12% | 5.5% |
| LMArena Expert | 1386 | 1435 |
| Humanity's Last Exam | 10.7% | — |
| SimpleQA Verified | — | 29.4% |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| GPQA (HELM) | 70.8% | — |
Multimodal GPT-5.4 mini leads
Claude Opus 4: 31.5 (#106), GPT-5.4 mini: 39.7 (#56)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1192 | 1245 |
| GeoBench | 49% | — |
| VPCT | 38% | — |
Multilingual GPT-5.4 mini leads
Claude Opus 4: 48.8 (#138), GPT-5.4 mini: 51.9 (#96)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1362 | 1405 |
| LMArena Chinese | 1386 | 1446 |
| LMArena French | 1372 | 1440 |
| LMArena German | 1391 | 1409 |
| LMArena Japanese | 1331 | 1374 |
| LMArena Korean | 1321 | 1368 |
| LMArena Russian | 1392 | 1417 |
| LMArena Spanish | 1389 | 1405 |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), GPT-5.4 mini: 74.1 (#102)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1406 | 1405 |
| IFEval | 91.8% | — |
Long Context GPT-5.4 mini leads
Claude Opus 4: 39.6 (#172), GPT-5.4 mini: 43.0 (#112)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1422 | 1407 |
| Fiction.LiveBench | 61.1% | — |
Writing & Preference GPT-5.4 mini leads
Claude Opus 4: 61.2 (#89), GPT-5.4 mini: 64.0 (#58)
| Benchmark | Claude Opus 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1377 | 1412 |
| LMArena Creative Writing | 1387 | 1370 |
| EQ-Bench Creative Writing | 1580 | 1665 |
| LMArena Multi-Turn | 1396 | 1429 |
| Short-Story Creative Writing | 83.6% | — |
| WildBench | 85.2% | — |
Frequently asked questions
Is Claude Opus 4 better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 43.1 on the Noometry Index.
Which is cheaper, Claude Opus 4 or GPT-5.4 mini?
GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or GPT-5.4 mini better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 45.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 200K.
How many benchmarks do Claude Opus 4 and GPT-5.4 mini share?
34 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and GPT-5.4 mini has 46.