Model comparison
Gemini 3 Pro vs GPT-4.1 mini
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 33.6 on the Noometry Index.
Last verified . 36 shared benchmarks.
GoogleSummary
- They share 36 benchmarks with published results for both. Gemini 3 Pro scores higher in 10 categories and GPT-4.1 mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 10.8.
- The biggest single-benchmark swing is ARC-AGI-1: 75% for Gemini 3 Pro and 3.5% for GPT-4.1 mini.
Side by side
| Gemini 3 Pro | GPT-4.1 mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 54.8 | 33.6 |
| Released | 2025-11-18 | 2025-04-14 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 33K |
| Input $ / M tokens | — | $0.40 |
| Output $ / M tokens | — | $1.60 |
| Results tracked | 67 | 47 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
Gemini 3 Pro: 51.6 (#39), GPT-4.1 mini: 30.6 (#293)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| SWE-bench Verified (bash only) | 74.2% | 23.9% |
| WeirdML | 69.9% | 37.6% |
| LMArena Coding | 1481 | 1367 |
| SWE-bench Verified | 72.9% | — |
| Aider Polyglot | — | 32.4% |
| LMArena WebDev | 1440 | — |
| SWE-bench Multilingual | 68.7% | — |
| SciCode | — | 40.4% |
| GSO | 18.6% | — |
| BigCodeBench Instruct | — | 48.9% |
| CadEval | — | 16% |
| ALE-Bench | 1,177 | — |
| AlgoTune | 1.83 | — |
Agentic & Tool Use Gemini 3 Pro leads
Gemini 3 Pro: 40.6 (#23), GPT-4.1 mini: 33.3 (#55)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 72.5% | 50.5% |
| Terminal-Bench | 69.4% | — |
| GDPval | 40.3% | — |
| Remote Labor Index | 1.3% | — |
| τ²-bench Airline | 80.5% | — |
| τ²-bench Banking | 18% | — |
| τ²-bench Retail | 75.9% | — |
| τ²-bench Telecom | 91% | — |
| DeepResearch Bench | 46.3% | — |
| BALROG | 58.1% | — |
| LMArena Search | 1207 | — |
| METR Time Horizons | 71% | — |
| Vending-Bench 2 | 5,478 | — |
Reasoning Gemini 3 Pro leads
Gemini 3 Pro: 52.5 (#31), GPT-4.1 mini: 10.8 (#340)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| ARC-AGI-2 | 31.1% | 0% |
| Kagi LLM Benchmark | 80.1% | 48.6% |
| ARC-AGI-1 | 75% | 3.5% |
| CritPt | 6.9% | 0% |
| Chess Puzzles | 31% | 7% |
| LMArena Hard Prompts | 1480 | 1349 |
| Epoch Capabilities Index | 152.92 | 135.01 |
| SimpleBench | 76.4% | — |
| NYT Connections (extended) | 94.4% | — |
| EnigmaEval | 18.2% | — |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 68.8% |
| LMCA | — | 21.1% |
| ForecastBench | 61.2 | — |
Math Gemini 3 Pro leads
Gemini 3 Pro: 49.9 (#59), GPT-4.1 mini: 24.1 (#270)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.4% | 44.7% |
| Omni-MATH | 55.5% | 49.1% |
| LMArena Math | 1476 | 1343 |
| FrontierMath (Feb 2025 set) | 37.6% | 4.5% |
| FrontierMath (Tiers 1-3) | — | 6.7% |
| MathArena Final-Answer Competitions | 67% | — |
| ProofBench | 20% | — |
| MATH Level 5 | — | 87.3% |
| FrontierMath Tier 4 (v1) | 18.8% | — |
Knowledge Gemini 3 Pro leads
Gemini 3 Pro: 64.4 (#16), GPT-4.1 mini: 34.7 (#194)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 92.6% | 65.8% |
| MMLU-Pro | 90.3% | 78.3% |
| GPQA (HELM) | 80.3% | 61.4% |
| LMArena Expert | 1475 | 1338 |
| Humanity's Last Exam | 37.5% | — |
| SimpleQA Verified | — | 12.7% |
| Vectara Hallucination Rate | 13.6% | — |
Multimodal Gemini 3 Pro leads
Gemini 3 Pro: 57.6 (#2), GPT-4.1 mini: 35.8 (#82)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | 1305 | 1181 |
| GeoBench | 84% | — |
| VPCT | 91% | — |
| LMArena Document | 1434 | — |
Multilingual Gemini 3 Pro leads
Gemini 3 Pro: 56.9 (#16), GPT-4.1 mini: 45.7 (#166)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1474 | 1318 |
| LMArena Chinese | 1523 | 1329 |
| LMArena French | 1492 | 1358 |
| LMArena German | 1515 | 1351 |
| LMArena Japanese | 1510 | 1290 |
| LMArena Korean | 1448 | 1298 |
| LMArena Russian | 1493 | 1324 |
| LMArena Spanish | 1470 | 1319 |
Instruction Following Gemini 3 Pro leads
Gemini 3 Pro: 76.3 (#45), GPT-4.1 mini: 73.7 (#118)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| IFEval | 87.7% | 90.4% |
| LMArena Instruction Following | 1458 | 1333 |
Long Context Gemini 3 Pro leads
Gemini 3 Pro: 44.0 (#79), GPT-4.1 mini: 31.8 (#275)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| LMArena Longer Query | 1471 | 1344 |
| Fiction.LiveBench | — | 44.4% |
| CL-bench | 15.8% | — |
Writing & Preference Gemini 3 Pro leads
Gemini 3 Pro: 66.4 (#35), GPT-4.1 mini: 48.6 (#199)
| Benchmark | Gemini 3 Pro | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1479 | 1340 |
| LMArena Creative Writing | 1482 | 1300 |
| EQ-Bench Creative Writing | 1525 | 1147 |
| WildBench | 85.9% | 83.8% |
| LMArena Multi-Turn | 1484 | 1354 |
Frequently asked questions
Is Gemini 3 Pro better than GPT-4.1 mini?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 33.6 on the Noometry Index.
Is Gemini 3 Pro or GPT-4.1 mini better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 51.6 versus 30.6 in the Noometry coding category.
How many benchmarks do Gemini 3 Pro and GPT-4.1 mini share?
36 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and GPT-4.1 mini has 47.