Model comparison
Claude Opus 4.1 vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 41.0 on the Noometry Index.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. Claude Opus 4.1 scores higher in 5 categories and GPT-5.4 mini in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 mini leads 45.5 to 22.3.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 12.6% for Claude Opus 4.1 and 51.2% 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 4.1.
- GPT-5.4 mini accepts more context: 400K tokens versus 200K.
Side by side
| Claude Opus 4.1 | GPT-5.4 mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 41.0 | 45.0 |
| Released | 2025-08-05 | 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 | 48 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Opus 4.1: 44.4 (#73), GPT-5.4 mini: 45.2 (#72)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| LMArena WebDev | 1390 | 1397 |
| WeirdML | 45.9% | 60.3% |
| LMArena Coding | 1479 | 1438 |
| ALE-Bench | 674.77 | 1,189 |
| SWE-bench Verified | 73.3% | — |
| FrontierCode | — | 27% |
| SciCode | — | 49.9% |
| AlgoTune | 1.34 | — |
Agentic & Tool Use Claude Opus 4.1 leads
Claude Opus 4.1: 35.0 (#41), GPT-5.4 mini: 29.9 (#81)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | 48.3% | 36.3% |
| Terminal-Bench | 38% | — |
| GDPval | 43.6% | — |
| Cybench | 42% | — |
| LMArena Search | 1148 | — |
| METR Time Horizons | 66.8% | — |
Reasoning Claude Opus 4.1 leads
Claude Opus 4.1: 32.2 (#76), GPT-5.4 mini: 30.4 (#85)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| Chess Puzzles | 7% | 24% |
| LMArena Hard Prompts | 1443 | 1424 |
| Mystery Game Puzzles | 21% | 11% |
| DTBench | 80% | 80% |
| LMCA | 37.1% | 40.8% |
| Epoch Capabilities Index | 144.12 | 148.84 |
| ForecastBench | 62 | 57 |
| ARC-AGI-2 | — | 18.9% |
| SimpleBench | 60% | — |
| Kagi LLM Benchmark | — | 37.9% |
| NYT Connections (extended) | — | 61.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | — | 10% |
| EnigmaEval | 7.2% | — |
| Thematic Generalization | — | 61.7% |
| EBR-Bench | 7.9% | — |
Math GPT-5.4 mini leads
Claude Opus 4.1: 22.3 (#277), GPT-5.4 mini: 45.5 (#75)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 12.6% | 51.2% |
| FrontierMath Tier 4 | 2.4% | 9.8% |
| OTIS Mock AIME 2024-2025 | 68.9% | 88.9% |
| LMArena Math | 1431 | 1419 |
| FrontierMath (Feb 2025 set) | 7.2% | 28.3% |
| FrontierMath Tier 4 (v1) | 4.2% | 2.1% |
| ProofBench | — | 21% |
Knowledge GPT-5.4 mini leads
Claude Opus 4.1: 42.0 (#101), GPT-5.4 mini: 51.5 (#67)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 77.3% | 86.9% |
| Vectara Hallucination Rate | 11.8% | 5.5% |
| LMArena Expert | 1439 | 1435 |
| Humanity's Last Exam | 11.5% | — |
| SimpleQA Verified | — | 29.4% |
| Confabulations | 17.1% | — |
Multimodal GPT-5.4 mini leads
Claude Opus 4.1: 26.8 (#119), GPT-5.4 mini: 39.7 (#56)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | — | 1245 |
| VPCT | 35% | — |
Multilingual Too close to call
Claude Opus 4.1: 52.0 (#95), GPT-5.4 mini: 51.9 (#96)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1405 | 1405 |
| LMArena Chinese | 1427 | 1446 |
| LMArena French | 1431 | 1440 |
| LMArena German | 1413 | 1409 |
| LMArena Japanese | 1378 | 1374 |
| LMArena Korean | 1380 | 1368 |
| LMArena Russian | 1422 | 1417 |
| LMArena Spanish | 1448 | 1405 |
Instruction Following Claude Opus 4.1 leads
Claude Opus 4.1: 75.6 (#58), GPT-5.4 mini: 74.1 (#102)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1435 | 1405 |
Long Context Claude Opus 4.1 leads
Claude Opus 4.1: 44.5 (#63), GPT-5.4 mini: 43.0 (#112)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1455 | 1407 |
Writing & Preference GPT-5.4 mini leads
Claude Opus 4.1: 62.4 (#74), GPT-5.4 mini: 64.0 (#58)
| Benchmark | Claude Opus 4.1 | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1419 | 1412 |
| LMArena Creative Writing | 1412 | 1370 |
| LMArena Multi-Turn | 1444 | 1429 |
| Short-Story Creative Writing | 84.7% | — |
| EQ-Bench Creative Writing | — | 1665 |
Frequently asked questions
Is Claude Opus 4.1 better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 41.0 on the Noometry Index.
Which is cheaper, Claude Opus 4.1 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 4.1 lists at $15 and $75.
Is Claude Opus 4.1 or GPT-5.4 mini better for coding?
They score almost the same on coding (44.4 vs 45.2); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 200K.
How many benchmarks do Claude Opus 4.1 and GPT-5.4 mini share?
34 benchmarks have published results for both models. Claude Opus 4.1 has 48 scored results on Noometry and GPT-5.4 mini has 46.