Model comparison
GPT-4.1 vs o4-mini
o4-mini is the stronger model overall, scoring 41.6 to 35.9 on the Noometry Index.
Last verified . 50 shared benchmarks.
Summary
- They share 50 benchmarks with published results for both. GPT-4.1 scores higher in 3 categories and o4-mini in 7 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where o4-mini leads 40.8 to 22.3.
- The biggest single-benchmark swing is ARC-AGI-1: 5.5% for GPT-4.1 and 58.7% for o4-mini.
- o4-mini is cheaper at $1.10 / $4.40 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 200K.
Side by side
| GPT-4.1 | o4-mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.9 | 41.6 |
| Released | 2025-04-14 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $2 | $1.10 |
| Output $ / M tokens | $8 | $4.40 |
| Results tracked | 52 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o4-mini leads
GPT-4.1: 34.4 (#238), o4-mini: 40.9 (#127)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| SWE-bench Verified (bash only) | 39.6% | 45% |
| Aider Polyglot | 52.4% | 72% |
| WeirdML | 39% | 52.6% |
| LMArena Coding | 1391 | 1368 |
| CadEval | 42% | 62% |
| ALE-Bench | 558.1 | 826.17 |
| SWE-bench Verified | 48.5% | — |
| GSO | — | 3.6% |
| AlgoTune | — | 1.72 |
Agentic & Tool Use GPT-4.1 leads
GPT-4.1: 34.7 (#43), o4-mini: 32.6 (#61)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | 53.2% |
| GDPval | — | 25.3% |
| METR Time Horizons | — | 63.9% |
Reasoning o4-mini leads
GPT-4.1: 11.7 (#339), o4-mini: 24.6 (#162)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| ARC-AGI-2 | 0.4% | 6.1% |
| SimpleBench | 27% | 38.7% |
| Kagi LLM Benchmark | 52.3% | 67.6% |
| ARC-AGI-1 | 5.5% | 58.7% |
| Chess Puzzles | 6% | 26% |
| EnigmaEval | 2.2% | 9.2% |
| LMArena Hard Prompts | 1384 | 1351 |
| DTBench | 68.3% | 77.6% |
| LMCA | 25.6% | 26.5% |
| Epoch Capabilities Index | 136.78 | 145.64 |
| ForecastBench | 61.5 | 61.8 |
| CritPt | — | 0.6% |
| Mystery Game Puzzles | — | 5% |
Math o4-mini leads
GPT-4.1: 22.3 (#280), o4-mini: 40.8 (#89)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 36.1% |
| OTIS Mock AIME 2024-2025 | 38.3% | 81.7% |
| Omni-MATH | 47.1% | 72% |
| LMArena Math | 1370 | 1389 |
| MATH Level 5 | 83% | 97.8% |
| FrontierMath (Feb 2025 set) | 5.5% | 24.8% |
| FrontierMath Tier 4 (v1) | 0% | 6.3% |
| FrontierMath Tier 4 | — | 4.9% |
Knowledge o4-mini leads
GPT-4.1: 37.1 (#160), o4-mini: 43.6 (#91)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| GPQA Diamond | 66.9% | 79.6% |
| Humanity's Last Exam | 5.4% | 18.1% |
| SimpleQA Verified | 31.1% | 19.6% |
| MMLU-Pro | 81.1% | 82% |
| Vectara Hallucination Rate | 5.6% | 18.6% |
| GPQA (HELM) | 65.9% | 73.5% |
| LMArena Expert | 1364 | 1343 |
| Confabulations | — | 15.8% |
Multimodal o4-mini leads
GPT-4.1: 38.2 (#67), o4-mini: 40.2 (#49)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| LMArena Vision | 1211 | 1194 |
| GeoBench | 72% | 64% |
| VPCT | — | 57.5% |
Multilingual GPT-4.1 leads
GPT-4.1: 49.4 (#133), o4-mini: 47.0 (#154)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| LMArena Non-English | 1370 | 1337 |
| LMArena Chinese | 1382 | 1354 |
| LMArena French | 1382 | 1364 |
| LMArena German | 1381 | 1336 |
| LMArena Japanese | 1319 | 1308 |
| LMArena Korean | 1339 | 1312 |
| LMArena Russian | 1377 | 1334 |
| LMArena Spanish | 1376 | 1347 |
Instruction Following o4-mini leads
GPT-4.1: 71.3 (#153), o4-mini: 75.2 (#68)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| IFEval | 83.8% | 92.8% |
| LMArena Instruction Following | 1367 | 1321 |
Long Context o4-mini leads
GPT-4.1: 40.0 (#163), o4-mini: 45.5 (#33)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| Fiction.LiveBench | 63.9% | 77.8% |
| LMArena Longer Query | 1385 | 1315 |
Writing & Preference GPT-4.1 leads
GPT-4.1: 57.6 (#125), o4-mini: 54.0 (#152)
| Benchmark | GPT-4.1 | o4-mini |
|---|---|---|
| LMArena Text | 1383 | 1353 |
| LMArena Creative Writing | 1363 | 1294 |
| WildBench | 85.4% | 85.4% |
| LMArena Multi-Turn | 1398 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 1420 | — |
Frequently asked questions
Is GPT-4.1 better than o4-mini?
o4-mini is the stronger model overall, scoring 41.6 to 35.9 on the Noometry Index.
Which is cheaper, GPT-4.1 or o4-mini?
o4-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GPT-4.1 or o4-mini better for coding?
o4-mini scores higher on coding benchmarks: 40.9 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 200K.
How many benchmarks do GPT-4.1 and o4-mini share?
50 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and o4-mini has 60.