Model comparison
GPT-4.1 mini vs o3-mini
o3-mini is the stronger model overall, scoring 36.7 to 33.5 on the Noometry Index. GPT-4.1 mini costs 2.8× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GPT-4.1 mini scores higher in 1 category and o3-mini in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where o3-mini leads 41.4 to 30.4.
- The biggest single-benchmark swing is CadEval: 16% for GPT-4.1 mini and 54% for o3-mini.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 200K.
Side by side
| GPT-4.1 mini | o3-mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 33.5 | 36.7 |
| Released | 2025-04-14 | 2024-12-20 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $0.40 | $1.10 |
| Output $ / M tokens | $1.60 | $4.40 |
| Results tracked | 47 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-mini leads
GPT-4.1 mini: 30.4 (#286), o3-mini: 41.4 (#116)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| Aider Polyglot | 32.4% | 60.4% |
| SciCode | 40.4% | 39.8% |
| WeirdML | 37.6% | 43.7% |
| LMArena Coding | 1367 | 1378 |
| CadEval | 16% | 54% |
| SWE-bench Verified (bash only) | 23.9% | — |
| GSO | — | 1.3% |
| BigCodeBench Instruct | 48.9% | — |
| LiveBench Coding | — | 82.7% |
Agentic & Tool Use GPT-4.1 mini leads
GPT-4.1 mini: 33.0 (#52), o3-mini: 29.3 (#79)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 50.5% | — |
| Cybench | — | 22.5% |
Reasoning o3-mini leads
GPT-4.1 mini: 10.9 (#335), o3-mini: 16.3 (#295)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| ARC-AGI-2 | 0% | 3% |
| ARC-AGI-1 | 3.5% | 34.5% |
| CritPt | 0% | 0.3% |
| Chess Puzzles | 7% | 17% |
| LMArena Hard Prompts | 1349 | 1366 |
| Mystery Game Puzzles | 7% | 7% |
| DTBench | 68.8% | 68.8% |
| LMCA | 21.1% | 19% |
| Epoch Capabilities Index | 135.01 | 140.35 |
| SimpleBench | — | 22.8% |
| Kagi LLM Benchmark | 48.6% | — |
| LiveBench Reasoning | — | 89.6% |
| LiveBench Data Analysis | — | 70.6% |
| ForecastBench | — | 59.6 |
| LiveBench | — | 75.9% |
Math o3-mini leads
GPT-4.1 mini: 23.1 (#264), o3-mini: 27.6 (#236)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6.7% | 18.6% |
| OTIS Mock AIME 2024-2025 | 44.7% | 76.9% |
| LMArena Math | 1343 | 1396 |
| MATH Level 5 | 87.3% | 96.5% |
| FrontierMath (Feb 2025 set) | 4.5% | 12.4% |
| FrontierMath Tier 4 | — | 0% |
| Omni-MATH | 49.1% | — |
| LiveBench Math | — | 77.3% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge o3-mini leads
GPT-4.1 mini: 34.7 (#183), o3-mini: 38.3 (#135)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| GPQA Diamond | 65.8% | 77% |
| SimpleQA Verified | 12.7% | 15.3% |
| LMArena Expert | 1338 | 1364 |
| MMLU-Pro | 78.3% | — |
| Confabulations | — | 17.9% |
| GPQA (HELM) | 61.4% | — |
Multimodal Not comparable
GPT-4.1 mini: 35.2 (#77), o3-mini: —
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| LMArena Vision | 1181 | — |
Multilingual Too close to call
GPT-4.1 mini: 45.9 (#161), o3-mini: 45.9 (#159)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| LMArena Non-English | 1318 | 1319 |
| LMArena Chinese | 1329 | 1379 |
| LMArena French | 1357 | 1333 |
| LMArena German | 1349 | 1303 |
| LMArena Japanese | 1291 | 1286 |
| LMArena Korean | 1298 | 1314 |
| LMArena Russian | 1324 | 1304 |
| LMArena Spanish | 1319 | 1321 |
Instruction Following o3-mini leads
GPT-4.1 mini: 73.9 (#113), o3-mini: 75.3 (#67)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1332 | 1338 |
| LiveBench Instruction Following | — | 84.4% |
| IFEval | 90.4% | — |
Long Context o3-mini leads
GPT-4.1 mini: 31.8 (#270), o3-mini: 33.8 (#251)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| Fiction.LiveBench | 44.4% | 50% |
| LMArena Longer Query | 1344 | 1343 |
Writing & Preference o3-mini leads
GPT-4.1 mini: 48.8 (#193), o3-mini: 50.4 (#177)
| Benchmark | GPT-4.1 mini | o3-mini |
|---|---|---|
| LMArena Text | 1341 | 1337 |
| LMArena Creative Writing | 1300 | 1286 |
| LMArena Multi-Turn | 1354 | 1320 |
| Short-Story Creative Writing | — | 61.7% |
| EQ-Bench Creative Writing | 1147 | — |
| WildBench | 83.8% | — |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is GPT-4.1 mini better than o3-mini?
o3-mini is the stronger model overall, scoring 36.7 to 33.5 on the Noometry Index. GPT-4.1 mini costs 2.8× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 mini or o3-mini?
GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is GPT-4.1 mini or o3-mini better for coding?
o3-mini scores higher on coding benchmarks: 41.4 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 GPT-4.1 mini and o3-mini share?
36 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and o3-mini has 51.