Model comparison
Claude 3.5 Sonnet vs o4-mini
o4-mini is the stronger model overall, scoring 41.6 to 34.6 on the Noometry Index.
Last verified . 43 shared benchmarks.
Summary
- They share 43 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 0 categories and o4-mini in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where o4-mini leads 40.8 to 19.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 81.7% for o4-mini.
Side by side
| Claude 3.5 Sonnet | o4-mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.6 | 41.6 |
| Released | 2024-06-20 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $1.10 |
| Output $ / M tokens | — | $4.40 |
| Results tracked | 60 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o4-mini leads
Claude 3.5 Sonnet: 39.0 (#165), o4-mini: 40.9 (#127)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| Aider Polyglot | 51.6% | 72% |
| GSO | 4.6% | 3.6% |
| WeirdML | 40% | 52.6% |
| LMArena Coding | 1342 | 1368 |
| CadEval | 48% | 62% |
| SWE-bench Verified (bash only) | — | 45% |
| BigCodeBench Instruct | 46.8% | — |
| LiveBench Coding | 67.1% | — |
| BigCodeBench Complete | 58.6% | — |
| ALE-Bench | — | 826.17 |
| AlgoTune | — | 1.72 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use Too close to call
Claude 3.5 Sonnet: 32.3 (#67), o4-mini: 32.6 (#61)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| METR Time Horizons | 45.2% | 63.9% |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| BALROG | 32.6% | — |
Reasoning o4-mini leads
Claude 3.5 Sonnet: 23.1 (#183), o4-mini: 24.6 (#162)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| SimpleBench | 41.4% | 38.7% |
| EnigmaEval | 0.9% | 9.2% |
| LMArena Hard Prompts | 1305 | 1351 |
| DTBench | 67.8% | 77.6% |
| Epoch Capabilities Index | 133.55 | 145.64 |
| ForecastBench | 60.7 | 61.8 |
| ARC-AGI-2 | — | 6.1% |
| Kagi LLM Benchmark | — | 67.6% |
| ARC-AGI-1 | — | 58.7% |
| CritPt | — | 0.6% |
| Chess Puzzles | — | 26% |
| LiveBench Reasoning | 56.7% | — |
| Mystery Game Puzzles | — | 5% |
| LiveBench Data Analysis | 55% | — |
| LMCA | — | 26.5% |
| LiveBench | 59% | — |
Math o4-mini leads
Claude 3.5 Sonnet: 19.2 (#288), o4-mini: 40.8 (#89)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 81.7% |
| Omni-MATH | 27.6% | 72% |
| LMArena Math | 1307 | 1389 |
| MATH Level 5 | 56.9% | 97.8% |
| FrontierMath (Feb 2025 set) | 2.1% | 24.8% |
| FrontierMath Tier 4 (v1) | 0% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| LiveBench Math | 52.3% | — |
Knowledge o4-mini leads
Claude 3.5 Sonnet: 28.6 (#245), o4-mini: 43.6 (#91)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| GPQA Diamond | 55.3% | 79.6% |
| Humanity's Last Exam | 4.1% | 18.1% |
| MMLU-Pro | 77.7% | 82% |
| Confabulations | 19.9% | 15.8% |
| GPQA (HELM) | 56.5% | 73.5% |
| LMArena Expert | 1265 | 1343 |
| SimpleQA Verified | — | 19.6% |
| Vectara Hallucination Rate | — | 18.6% |
| MMLU | 87.3% | — |
Multimodal o4-mini leads
Claude 3.5 Sonnet: 26.5 (#120), o4-mini: 40.2 (#49)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| LMArena Vision | 1125 | 1194 |
| GeoBench | 62% | 64% |
| VPCT | 33% | 57.5% |
| Video-MME | 60% | — |
Multilingual o4-mini leads
Claude 3.5 Sonnet: 43.2 (#185), o4-mini: 47.0 (#154)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| LMArena Non-English | 1283 | 1337 |
| LMArena Chinese | 1272 | 1354 |
| LMArena French | 1305 | 1364 |
| LMArena German | 1297 | 1336 |
| LMArena Japanese | 1234 | 1308 |
| LMArena Korean | 1200 | 1312 |
| LMArena Russian | 1306 | 1334 |
| LMArena Spanish | 1290 | 1347 |
Instruction Following o4-mini leads
Claude 3.5 Sonnet: 68.8 (#182), o4-mini: 75.2 (#68)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| IFEval | 85.5% | 92.8% |
| LMArena Instruction Following | 1297 | 1321 |
| LiveBench Instruction Following | 69.3% | — |
Long Context o4-mini leads
Claude 3.5 Sonnet: 39.9 (#167), o4-mini: 45.5 (#33)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| LMArena Longer Query | 1311 | 1315 |
| Fiction.LiveBench | — | 77.8% |
Writing & Preference o4-mini leads
Claude 3.5 Sonnet: 52.9 (#164), o4-mini: 54.0 (#152)
| Benchmark | Claude 3.5 Sonnet | o4-mini |
|---|---|---|
| LMArena Text | 1298 | 1353 |
| LMArena Creative Writing | 1292 | 1294 |
| Short-Story Creative Writing | 80.3% | 75% |
| WildBench | 79.2% | 85.4% |
| LMArena Multi-Turn | 1326 | 1350 |
| EQ-Bench Creative Writing | 1451 | — |
| LiveBench Language | 53.8% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than o4-mini?
o4-mini is the stronger model overall, scoring 41.6 to 34.6 on the Noometry Index.
Is Claude 3.5 Sonnet or o4-mini better for coding?
o4-mini scores higher on coding benchmarks: 40.9 versus 39.0 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and o4-mini share?
43 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and o4-mini has 60.