Model comparison
GPT-4.1 mini vs o3
o3 is the stronger model overall, scoring 47.5 to 33.6 on the Noometry Index. GPT-4.1 mini costs 5.0× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 45 shared benchmarks.
Summary
- They share 45 benchmarks with published results for both. GPT-4.1 mini scores higher in 1 category and o3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where o3 leads 50.2 to 24.1.
- The biggest single-benchmark swing is CadEval: 16% for GPT-4.1 mini and 74% for o3.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $2 / $8 for o3.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 200K.
Side by side
| GPT-4.1 mini | o3 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 33.6 | 47.5 |
| Released | 2025-04-14 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $0.40 | $2 |
| Output $ / M tokens | $1.60 | $8 |
| Results tracked | 47 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
GPT-4.1 mini: 30.6 (#293), o3: 46.8 (#64)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| SWE-bench Verified (bash only) | 23.9% | 58.4% |
| Aider Polyglot | 32.4% | 81.3% |
| WeirdML | 37.6% | 52.4% |
| LMArena Coding | 1367 | 1408 |
| CadEval | 16% | 74% |
| SWE-bench Verified | — | 62.3% |
| SciCode | 40.4% | — |
| GSO | — | 8.8% |
| BigCodeBench Instruct | 48.9% | — |
| ALE-Bench | — | 933.55 |
Agentic & Tool Use o3 leads
GPT-4.1 mini: 33.3 (#55), o3: 34.5 (#44)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 50.5% | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
GPT-4.1 mini: 10.8 (#340), o3: 32.0 (#78)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| ARC-AGI-2 | 0% | 6.5% |
| Kagi LLM Benchmark | 48.6% | 67.6% |
| ARC-AGI-1 | 3.5% | 60.8% |
| CritPt | 0% | 1.4% |
| Chess Puzzles | 7% | 38% |
| LMArena Hard Prompts | 1349 | 1402 |
| Mystery Game Puzzles | 7% | 29% |
| DTBench | 68.8% | 84.8% |
| LMCA | 21.1% | 39.7% |
| Epoch Capabilities Index | 135.01 | 146.86 |
| SimpleBench | — | 53.1% |
| EnigmaEval | — | 13.1% |
| ForecastBench | — | 62.5 |
Math o3 leads
GPT-4.1 mini: 24.1 (#270), o3: 50.2 (#58)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6.7% | 33.3% |
| OTIS Mock AIME 2024-2025 | 44.7% | 84.4% |
| Omni-MATH | 49.1% | 71.4% |
| LMArena Math | 1343 | 1426 |
| MATH Level 5 | 87.3% | 97.8% |
| FrontierMath (Feb 2025 set) | 4.5% | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
GPT-4.1 mini: 34.7 (#194), o3: 54.6 (#52)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| GPQA Diamond | 65.8% | 81.8% |
| SimpleQA Verified | 12.7% | 49.4% |
| MMLU-Pro | 78.3% | 85.9% |
| GPQA (HELM) | 61.4% | 75.3% |
| LMArena Expert | 1338 | 1402 |
| Humanity's Last Exam | — | 20.3% |
| Confabulations | — | 14.4% |
Multimodal o3 leads
GPT-4.1 mini: 35.8 (#82), o3: 41.4 (#36)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| LMArena Vision | 1181 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual o3 leads
GPT-4.1 mini: 45.7 (#166), o3: 51.7 (#105)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| LMArena Non-English | 1318 | 1401 |
| LMArena Chinese | 1329 | 1437 |
| LMArena French | 1358 | 1430 |
| LMArena German | 1351 | 1420 |
| LMArena Japanese | 1290 | 1403 |
| LMArena Korean | 1298 | 1370 |
| LMArena Russian | 1324 | 1406 |
| LMArena Spanish | 1319 | 1395 |
Instruction Following Too close to call
GPT-4.1 mini: 73.7 (#118), o3: 72.8 (#127)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| IFEval | 90.4% | 86.9% |
| LMArena Instruction Following | 1333 | 1368 |
Long Context o3 leads
GPT-4.1 mini: 31.8 (#275), o3: 53.3 (#6)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| Fiction.LiveBench | 44.4% | 88.9% |
| LMArena Longer Query | 1344 | 1372 |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
GPT-4.1 mini: 48.6 (#199), o3: 63.5 (#64)
| Benchmark | GPT-4.1 mini | o3 |
|---|---|---|
| LMArena Text | 1340 | 1410 |
| LMArena Creative Writing | 1300 | 1359 |
| EQ-Bench Creative Writing | 1147 | 1676 |
| WildBench | 83.8% | 86.1% |
| LMArena Multi-Turn | 1354 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
Frequently asked questions
Is GPT-4.1 mini better than o3?
o3 is the stronger model overall, scoring 47.5 to 33.6 on the Noometry Index. GPT-4.1 mini costs 5.0× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 mini or o3?
GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; o3 lists at $2 and $8.
Is GPT-4.1 mini or o3 better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 30.6 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 share?
45 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and o3 has 63.