Model comparison
Claude Sonnet 4 vs GPT-4.1 mini
Claude Sonnet 4 is the stronger model overall, scoring 40.6 to 33.5 on the Noometry Index. GPT-4.1 mini costs 8.6× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Last verified . 40 shared benchmarks.
Summary
- They share 40 benchmarks with published results for both. Claude Sonnet 4 scores higher in 8 categories and GPT-4.1 mini in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 4 leads 41.5 to 23.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 64.9% for Claude Sonnet 4 and 23.9% for GPT-4.1 mini.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 200K.
Side by side
| Claude Sonnet 4 | GPT-4.1 mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 40.6 | 33.5 |
| Released | 2025-05-22 | 2025-04-14 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 64K | 33K |
| Input $ / M tokens | $3 | $0.40 |
| Output $ / M tokens | $15 | $1.60 |
| Results tracked | 58 | 47 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 44.0 (#70), GPT-4.1 mini: 30.4 (#286)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| SWE-bench Verified (bash only) | 64.9% | 23.9% |
| Aider Polyglot | 61.3% | 32.4% |
| SciCode | 40% | 40.4% |
| WeirdML | 46.1% | 37.6% |
| LMArena Coding | 1414 | 1367 |
| GSO | 4.9% | — |
| BigCodeBench Instruct | — | 48.9% |
| CadEval | — | 16% |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.3 (#29), GPT-4.1 mini: 33.0 (#52)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 50.5% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Claude Sonnet 4 leads
Claude Sonnet 4: 23.0 (#169), GPT-4.1 mini: 10.9 (#335)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| ARC-AGI-2 | 5.9% | 0% |
| Kagi LLM Benchmark | 73% | 48.6% |
| ARC-AGI-1 | 40% | 3.5% |
| CritPt | 0.3% | 0% |
| LMArena Hard Prompts | 1372 | 1349 |
| DTBench | 77.1% | 68.8% |
| LMCA | 29% | 21.1% |
| Epoch Capabilities Index | 141.69 | 135.01 |
| SimpleBench | 45.5% | — |
| Chess Puzzles | — | 7% |
| EnigmaEval | 3.1% | — |
| Mystery Game Puzzles | — | 7% |
| ForecastBench | 60.2 | — |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 41.5 (#71), GPT-4.1 mini: 23.1 (#264)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 44.7% |
| Omni-MATH | 60.2% | 49.1% |
| LMArena Math | 1375 | 1343 |
| MATH Level 5 | 84.4% | 87.3% |
| FrontierMath (Feb 2025 set) | 4.1% | 4.5% |
| FrontierMath (Tiers 1-3) | — | 6.7% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Claude Sonnet 4 leads
Claude Sonnet 4: 41.6 (#99), GPT-4.1 mini: 34.7 (#183)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 79.2% | 65.8% |
| MMLU-Pro | 84.3% | 78.3% |
| GPQA (HELM) | 70.6% | 61.4% |
| LMArena Expert | 1371 | 1338 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 12.7% |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
Multimodal GPT-4.1 mini leads
Claude Sonnet 4: 25.6 (#116), GPT-4.1 mini: 35.2 (#77)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | 1191 | 1181 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Claude Sonnet 4 leads
Claude Sonnet 4: 46.9 (#151), GPT-4.1 mini: 45.9 (#161)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1333 | 1318 |
| LMArena Chinese | 1351 | 1329 |
| LMArena French | 1364 | 1357 |
| LMArena German | 1331 | 1349 |
| LMArena Japanese | 1301 | 1291 |
| LMArena Korean | 1293 | 1298 |
| LMArena Russian | 1355 | 1324 |
| LMArena Spanish | 1356 | 1319 |
Instruction Following GPT-4.1 mini leads
Claude Sonnet 4: 71.8 (#141), GPT-4.1 mini: 73.9 (#113)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| IFEval | 84% | 90.4% |
| LMArena Instruction Following | 1375 | 1332 |
Long Context Claude Sonnet 4 leads
Claude Sonnet 4: 33.6 (#254), GPT-4.1 mini: 31.8 (#270)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| Fiction.LiveBench | 46.9% | 44.4% |
| LMArena Longer Query | 1398 | 1344 |
Writing & Preference Claude Sonnet 4 leads
Claude Sonnet 4: 57.3 (#127), GPT-4.1 mini: 48.8 (#193)
| Benchmark | Claude Sonnet 4 | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1351 | 1341 |
| LMArena Creative Writing | 1345 | 1300 |
| EQ-Bench Creative Writing | 1483 | 1147 |
| WildBench | 83.8% | 83.8% |
| LMArena Multi-Turn | 1376 | 1354 |
| Short-Story Creative Writing | 81.4% | — |
Frequently asked questions
Is Claude Sonnet 4 better than GPT-4.1 mini?
Claude Sonnet 4 is the stronger model overall, scoring 40.6 to 33.5 on the Noometry Index. GPT-4.1 mini costs 8.6× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 4 or GPT-4.1 mini?
GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or GPT-4.1 mini better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 44.0 versus 30.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 mini does, with 1.05M tokens against 200K.
How many benchmarks do Claude Sonnet 4 and GPT-4.1 mini share?
40 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and GPT-4.1 mini has 47.