Model comparison
Claude 3.5 Sonnet vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 34.4 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 1 category and GPT-5.4 mini in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 mini leads 45.4 to 18.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 88.9% for GPT-5.4 mini.
Side by side
| Claude 3.5 Sonnet | GPT-5.4 mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.4 | 44.8 |
| Released | 2024-06-20 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.75 |
| Output $ / M tokens | — | $4.50 |
| Results tracked | 60 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
Claude 3.5 Sonnet: 39.5 (#145), GPT-5.4 mini: 45.0 (#67)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| WeirdML | 40% | 60.3% |
| LMArena Coding | 1343 | 1438 |
| FrontierCode | — | 27% |
| Aider Polyglot | 51.6% | — |
| LMArena WebDev | — | 1397 |
| SciCode | — | 49.9% |
| GSO | 4.6% | — |
| BigCodeBench Instruct | 46.8% | — |
| LiveBench Coding | 67.1% | — |
| BigCodeBench Complete | 58.6% | — |
| CadEval | 48% | — |
| ALE-Bench | — | 1,189 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 32.1 (#61), GPT-5.4 mini: 29.6 (#74)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| DeepResearch Bench | — | 36.3% |
| BALROG | 32.6% | — |
| METR Time Horizons | 45.2% | — |
Reasoning GPT-5.4 mini leads
Claude 3.5 Sonnet: 22.5 (#176), GPT-5.4 mini: 30.3 (#77)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| LMArena Hard Prompts | 1305 | 1424 |
| DTBench | 67.8% | 80% |
| Epoch Capabilities Index | 133.55 | 148.83 |
| ForecastBench | 60.7 | 57 |
| ARC-AGI-2 | — | 18.9% |
| SimpleBench | 41.4% | — |
| Kagi LLM Benchmark | — | 37.9% |
| NYT Connections (extended) | — | 61.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | — | 10% |
| Chess Puzzles | — | 24% |
| EnigmaEval | 0.9% | — |
| Thematic Generalization | — | 61.7% |
| LiveBench Reasoning | 56.7% | — |
| Mystery Game Puzzles | — | 11% |
| LiveBench Data Analysis | 55% | — |
| LMCA | — | 40.8% |
| LiveBench | 59% | — |
Math GPT-5.4 mini leads
Claude 3.5 Sonnet: 18.0 (#282), GPT-5.4 mini: 45.4 (#63)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 88.9% |
| LMArena Math | 1306 | 1418 |
| FrontierMath (Feb 2025 set) | 2.1% | 28.3% |
| FrontierMath Tier 4 (v1) | 0% | 2.1% |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| ProofBench | — | 21% |
| Omni-MATH | 27.6% | — |
| LiveBench Math | 52.3% | — |
| MATH Level 5 | 56.9% | — |
Knowledge GPT-5.4 mini leads
Claude 3.5 Sonnet: 28.5 (#236), GPT-5.4 mini: 51.4 (#62)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 55.3% | 86.9% |
| LMArena Expert | 1264 | 1434 |
| Humanity's Last Exam | 4.1% | — |
| SimpleQA Verified | — | 29.4% |
| MMLU-Pro | 77.7% | — |
| Confabulations | 19.9% | — |
| Vectara Hallucination Rate | — | 5.5% |
| GPQA (HELM) | 56.5% | — |
| MMLU | 87.3% | — |
Multimodal GPT-5.4 mini leads
Claude 3.5 Sonnet: 26.0 (#115), GPT-5.4 mini: 39.0 (#51)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1125 | 1245 |
| Video-MME | 60% | — |
| GeoBench | 62% | — |
| VPCT | 33% | — |
Multilingual GPT-5.4 mini leads
Claude 3.5 Sonnet: 43.4 (#180), GPT-5.4 mini: 52.1 (#94)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1283 | 1405 |
| LMArena Chinese | 1271 | 1446 |
| LMArena French | 1304 | 1440 |
| LMArena German | 1297 | 1407 |
| LMArena Japanese | 1234 | 1375 |
| LMArena Korean | 1200 | 1369 |
| LMArena Russian | 1305 | 1417 |
| LMArena Spanish | 1289 | 1403 |
Instruction Following GPT-5.4 mini leads
Claude 3.5 Sonnet: 69.0 (#177), GPT-5.4 mini: 74.3 (#99)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1297 | 1405 |
| LiveBench Instruction Following | 69.3% | — |
| IFEval | 85.5% | — |
Long Context GPT-5.4 mini leads
Claude 3.5 Sonnet: 39.9 (#162), GPT-5.4 mini: 42.9 (#107)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1311 | 1407 |
Writing & Preference GPT-5.4 mini leads
Claude 3.5 Sonnet: 53.0 (#159), GPT-5.4 mini: 64.2 (#56)
| Benchmark | Claude 3.5 Sonnet | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1298 | 1411 |
| LMArena Creative Writing | 1292 | 1370 |
| EQ-Bench Creative Writing | 1451 | 1665 |
| LMArena Multi-Turn | 1326 | 1429 |
| Short-Story Creative Writing | 80.3% | — |
| WildBench | 79.2% | — |
| LiveBench Language | 53.8% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 34.4 on the Noometry Index.
Is Claude 3.5 Sonnet or GPT-5.4 mini better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 39.5 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and GPT-5.4 mini share?
27 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and GPT-5.4 mini has 46.