Model comparison
Gemini 3 Pro vs GPT-5 Mini
Gemini 3 Pro is the stronger model overall, scoring 54.7 to 41.7 on the Noometry Index.
Last verified . 49 shared benchmarks.
GoogleSummary
- They share 49 benchmarks with published results for both. Gemini 3 Pro scores higher in 10 categories and GPT-5 Mini in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 24.1.
- The biggest single-benchmark swing is VPCT: 91% for Gemini 3 Pro and 40.2% for GPT-5 Mini.
Side by side
| Gemini 3 Pro | GPT-5 Mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 54.7 | 41.7 |
| Released | 2025-11-18 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $2 |
| Results tracked | 67 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
Gemini 3 Pro: 52.1 (#33), GPT-5 Mini: 40.0 (#134)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| SWE-bench Verified | 72.9% | 64.7% |
| SWE-bench Verified (bash only) | 74.2% | 59.8% |
| SWE-bench Multilingual | 68.7% | 39.7% |
| WeirdML | 69.9% | 52.7% |
| LMArena Coding | 1482 | 1406 |
| ALE-Bench | 1,177 | 799.77 |
| AlgoTune | 1.83 | 1.38 |
| LMArena WebDev | 1439 | — |
| SciCode | — | 39.2% |
| GSO | 18.6% | — |
Agentic & Tool Use Gemini 3 Pro leads
Gemini 3 Pro: 40.5 (#21), GPT-5 Mini: 31.0 (#64)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | 69.4% | 34.8% |
| Berkeley Function Calling Leaderboard | 72.5% | 55.5% |
| Vending-Bench 2 | 5,478 | -31.18 |
| 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% | — |
Reasoning Gemini 3 Pro leads
Gemini 3 Pro: 52.5 (#28), GPT-5 Mini: 24.1 (#153)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 31.1% | 4.4% |
| Kagi LLM Benchmark | 80.1% | 70.3% |
| ARC-AGI-1 | 75% | 54.3% |
| CritPt | 6.9% | 0% |
| Chess Puzzles | 31% | 30% |
| EnigmaEval | 18.2% | 8.2% |
| LMArena Hard Prompts | 1481 | 1380 |
| Epoch Capabilities Index | 152.96 | 145.51 |
| ForecastBench | 61.2 | 61 |
| SimpleBench | 76.4% | — |
| NYT Connections (extended) | 94.4% | — |
| Mystery Game Puzzles | — | 10% |
| DTBench | — | 80.5% |
| LMCA | — | 34.2% |
Math Gemini 3 Pro leads
Gemini 3 Pro: 48.8 (#50), GPT-5 Mini: 46.3 (#59)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.4% | 86.7% |
| ProofBench | 20% | 9% |
| Omni-MATH | 55.5% | 72.2% |
| LMArena Math | 1475 | 1377 |
| FrontierMath (Feb 2025 set) | 37.6% | 27.2% |
| FrontierMath Tier 4 (v1) | 18.8% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| MathArena Final-Answer Competitions | 67% | — |
| MATH Level 5 | — | 97.8% |
Knowledge Gemini 3 Pro leads
Gemini 3 Pro: 64.3 (#15), GPT-5 Mini: 45.5 (#80)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 92.6% | 75% |
| Humanity's Last Exam | 37.5% | 19.4% |
| MMLU-Pro | 90.3% | 83.5% |
| Vectara Hallucination Rate | 13.6% | 12.9% |
| GPQA (HELM) | 80.3% | 75.6% |
| LMArena Expert | 1475 | 1377 |
| SimpleQA Verified | — | 21.6% |
| Confabulations | — | 13.3% |
Multimodal Gemini 3 Pro leads
Gemini 3 Pro: 56.9 (#1), GPT-5 Mini: 35.0 (#80)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| LMArena Vision | 1305 | 1202 |
| VPCT | 91% | 40.2% |
| GeoBench | 84% | — |
| LMArena Document | 1434 | — |
Multilingual Gemini 3 Pro leads
Gemini 3 Pro: 57.1 (#14), GPT-5 Mini: 49.1 (#131)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1474 | 1363 |
| LMArena Chinese | 1522 | 1387 |
| LMArena French | 1492 | 1385 |
| LMArena German | 1517 | 1366 |
| LMArena Japanese | 1513 | 1342 |
| LMArena Korean | 1449 | 1306 |
| LMArena Russian | 1493 | 1362 |
| LMArena Spanish | 1468 | 1355 |
Instruction Following Too close to call
Gemini 3 Pro: 76.4 (#42), GPT-5 Mini: 76.4 (#43)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| IFEval | 87.7% | 92.7% |
| LMArena Instruction Following | 1458 | 1357 |
Long Context Gemini 3 Pro leads
Gemini 3 Pro: 44.0 (#75), GPT-5 Mini: 41.9 (#127)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1472 | 1355 |
| Fiction.LiveBench | — | 69.4% |
| CL-bench | 15.8% | — |
Writing & Preference Gemini 3 Pro leads
Gemini 3 Pro: 66.6 (#34), GPT-5 Mini: 55.4 (#143)
| Benchmark | Gemini 3 Pro | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1479 | 1373 |
| LMArena Creative Writing | 1482 | 1325 |
| EQ-Bench Creative Writing | 1525 | 1313 |
| WildBench | 85.9% | 85.5% |
| LMArena Multi-Turn | 1484 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
Frequently asked questions
Is Gemini 3 Pro better than GPT-5 Mini?
Gemini 3 Pro is the stronger model overall, scoring 54.7 to 41.7 on the Noometry Index.
Is Gemini 3 Pro or GPT-5 Mini better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 52.1 versus 40.0 in the Noometry coding category.
How many benchmarks do Gemini 3 Pro and GPT-5 Mini share?
49 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and GPT-5 Mini has 60.