Model comparison
GPT-4.1 vs GPT-5 Mini
GPT-5 Mini is the stronger model overall, scoring 41.7 to 35.7 on the Noometry Index.
Last verified . 48 shared benchmarks.
Summary
- They share 48 benchmarks with published results for both. GPT-4.1 scores higher in 4 categories and GPT-5 Mini in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 21.4.
- The biggest single-benchmark swing is ARC-AGI-1: 5.5% for GPT-4.1 and 54.3% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-4.1 | GPT-5 Mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.7 | 41.7 |
| Released | 2025-04-14 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 33K | 128K |
| Input $ / M tokens | $2 | $0.25 |
| Output $ / M tokens | $8 | $2 |
| Results tracked | 52 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-4.1: 34.2 (#228), GPT-5 Mini: 40.0 (#134)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| SWE-bench Verified | 48.5% | 64.7% |
| SWE-bench Verified (bash only) | 39.6% | 59.8% |
| WeirdML | 39% | 52.7% |
| LMArena Coding | 1391 | 1406 |
| ALE-Bench | 558.1 | 799.77 |
| Aider Polyglot | 52.4% | — |
| SWE-bench Multilingual | — | 39.7% |
| SciCode | — | 39.2% |
| CadEval | 42% | — |
| AlgoTune | — | 1.38 |
Agentic & Tool Use GPT-4.1 leads
GPT-4.1: 34.4 (#40), GPT-5 Mini: 31.0 (#64)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | 55.5% |
| Terminal-Bench | — | 34.8% |
| Vending-Bench 2 | — | -31.18 |
Reasoning GPT-5 Mini leads
GPT-4.1: 11.7 (#334), GPT-5 Mini: 24.1 (#153)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 0.4% | 4.4% |
| Kagi LLM Benchmark | 52.3% | 70.3% |
| ARC-AGI-1 | 5.5% | 54.3% |
| Chess Puzzles | 6% | 30% |
| EnigmaEval | 2.2% | 8.2% |
| LMArena Hard Prompts | 1384 | 1380 |
| DTBench | 68.3% | 80.5% |
| LMCA | 25.6% | 34.2% |
| Epoch Capabilities Index | 136.79 | 145.51 |
| ForecastBench | 61.5 | 61 |
| SimpleBench | 27% | — |
| CritPt | — | 0% |
| Mystery Game Puzzles | — | 10% |
Math GPT-5 Mini leads
GPT-4.1: 21.4 (#269), GPT-5 Mini: 46.3 (#59)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 46.7% |
| OTIS Mock AIME 2024-2025 | 38.3% | 86.7% |
| Omni-MATH | 47.1% | 72.2% |
| LMArena Math | 1369 | 1377 |
| MATH Level 5 | 83% | 97.8% |
| FrontierMath (Feb 2025 set) | 5.5% | 27.2% |
| FrontierMath Tier 4 (v1) | 0% | 6.3% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
Knowledge GPT-5 Mini leads
GPT-4.1: 37.0 (#152), GPT-5 Mini: 45.5 (#80)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 66.9% | 75% |
| Humanity's Last Exam | 5.4% | 19.4% |
| SimpleQA Verified | 31.1% | 21.6% |
| MMLU-Pro | 81.1% | 83.5% |
| Vectara Hallucination Rate | 5.6% | 12.9% |
| GPQA (HELM) | 65.9% | 75.6% |
| LMArena Expert | 1364 | 1377 |
| Confabulations | — | 13.3% |
Multimodal GPT-4.1 leads
GPT-4.1: 37.5 (#66), GPT-5 Mini: 35.0 (#80)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| LMArena Vision | 1210 | 1202 |
| GeoBench | 72% | — |
| VPCT | — | 40.2% |
Multilingual Too close to call
GPT-4.1: 49.6 (#128), GPT-5 Mini: 49.1 (#131)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1370 | 1363 |
| LMArena Chinese | 1382 | 1387 |
| LMArena French | 1383 | 1385 |
| LMArena German | 1380 | 1366 |
| LMArena Japanese | 1320 | 1342 |
| LMArena Korean | 1339 | 1306 |
| LMArena Russian | 1377 | 1362 |
| LMArena Spanish | 1375 | 1355 |
Instruction Following GPT-5 Mini leads
GPT-4.1: 71.5 (#148), GPT-5 Mini: 76.4 (#43)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| IFEval | 83.8% | 92.7% |
| LMArena Instruction Following | 1367 | 1357 |
Long Context GPT-5 Mini leads
GPT-4.1: 40.0 (#158), GPT-5 Mini: 41.9 (#127)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| Fiction.LiveBench | 63.9% | 69.4% |
| LMArena Longer Query | 1385 | 1355 |
Writing & Preference GPT-4.1 leads
GPT-4.1: 57.8 (#120), GPT-5 Mini: 55.4 (#143)
| Benchmark | GPT-4.1 | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1383 | 1373 |
| LMArena Creative Writing | 1363 | 1325 |
| EQ-Bench Creative Writing | 1420 | 1313 |
| WildBench | 85.4% | 85.5% |
| LMArena Multi-Turn | 1398 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
Frequently asked questions
Is GPT-4.1 better than GPT-5 Mini?
GPT-5 Mini is the stronger model overall, scoring 41.7 to 35.7 on the Noometry Index.
Which is cheaper, GPT-4.1 or GPT-5 Mini?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GPT-4.1 or GPT-5 Mini better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.0 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 400K.
How many benchmarks do GPT-4.1 and GPT-5 Mini share?
48 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and GPT-5 Mini has 60.