Model comparison
Claude Sonnet 4 vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 40.6 on the Noometry Index.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. Claude Sonnet 4 scores higher in 1 category and GPT-5.4 mini in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where GPT-5.4 mini leads 39.0 to 25.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 73% for Claude Sonnet 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 $3 / $15 for Claude Sonnet 4.
- GPT-5.4 mini accepts more context: 400K tokens versus 200K.
Side by side
| Claude Sonnet 4 | GPT-5.4 mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 40.6 | 44.8 |
| Released | 2025-05-22 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 400K |
| Max output | 64K | 128K |
| Input $ / M tokens | $3 | $0.75 |
| Output $ / M tokens | $15 | $4.50 |
| Results tracked | 58 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Sonnet 4: 44.0 (#70), GPT-5.4 mini: 45.0 (#67)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| SciCode | 40% | 49.9% |
| WeirdML | 46.1% | 60.3% |
| LMArena Coding | 1414 | 1438 |
| ALE-Bench | 655.35 | 1,189 |
| FrontierCode | — | 27% |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1397 |
| GSO | 4.9% | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.3 (#29), GPT-5.4 mini: 29.6 (#74)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | 46.6% | 36.3% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning GPT-5.4 mini leads
Claude Sonnet 4: 23.0 (#169), GPT-5.4 mini: 30.3 (#77)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| ARC-AGI-2 | 5.9% | 18.9% |
| Kagi LLM Benchmark | 73% | 37.9% |
| ARC-AGI-1 | 40% | 63.7% |
| CritPt | 0.3% | 10% |
| LMArena Hard Prompts | 1372 | 1424 |
| DTBench | 77.1% | 80% |
| LMCA | 29% | 40.8% |
| Epoch Capabilities Index | 141.69 | 148.83 |
| ForecastBench | 60.2 | 57 |
| SimpleBench | 45.5% | — |
| NYT Connections (extended) | — | 61.8% |
| Chess Puzzles | — | 24% |
| EnigmaEval | 3.1% | — |
| Thematic Generalization | — | 61.7% |
| Mystery Game Puzzles | — | 11% |
Math GPT-5.4 mini leads
Claude Sonnet 4: 41.5 (#71), GPT-5.4 mini: 45.4 (#63)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 88.9% |
| LMArena Math | 1375 | 1418 |
| FrontierMath (Feb 2025 set) | 4.1% | 28.3% |
| FrontierMath Tier 4 (v1) | 0% | 2.1% |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| ProofBench | — | 21% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
Knowledge GPT-5.4 mini leads
Claude Sonnet 4: 41.6 (#99), GPT-5.4 mini: 51.4 (#62)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 79.2% | 86.9% |
| Vectara Hallucination Rate | 10.3% | 5.5% |
| LMArena Expert | 1371 | 1434 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 29.4% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal GPT-5.4 mini leads
Claude Sonnet 4: 25.6 (#116), GPT-5.4 mini: 39.0 (#51)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1191 | 1245 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual GPT-5.4 mini leads
Claude Sonnet 4: 46.9 (#151), GPT-5.4 mini: 52.1 (#94)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1333 | 1405 |
| LMArena Chinese | 1351 | 1446 |
| LMArena French | 1364 | 1440 |
| LMArena German | 1331 | 1407 |
| LMArena Japanese | 1301 | 1375 |
| LMArena Korean | 1293 | 1369 |
| LMArena Russian | 1355 | 1417 |
| LMArena Spanish | 1356 | 1403 |
Instruction Following GPT-5.4 mini leads
Claude Sonnet 4: 71.8 (#141), GPT-5.4 mini: 74.3 (#99)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1375 | 1405 |
| IFEval | 84% | — |
Long Context GPT-5.4 mini leads
Claude Sonnet 4: 33.6 (#254), GPT-5.4 mini: 42.9 (#107)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1398 | 1407 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference GPT-5.4 mini leads
Claude Sonnet 4: 57.3 (#127), GPT-5.4 mini: 64.2 (#56)
| Benchmark | Claude Sonnet 4 | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1351 | 1411 |
| LMArena Creative Writing | 1345 | 1370 |
| EQ-Bench Creative Writing | 1483 | 1665 |
| LMArena Multi-Turn | 1376 | 1429 |
| Short-Story Creative Writing | 81.4% | — |
| WildBench | 83.8% | — |
Frequently asked questions
Is Claude Sonnet 4 better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 40.6 on the Noometry Index.
Which is cheaper, Claude Sonnet 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 Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or GPT-5.4 mini better for coding?
They score almost the same on coding (44.0 vs 45.0); 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 Sonnet 4 and GPT-5.4 mini share?
36 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and GPT-5.4 mini has 46.