Model comparison
Gemini 3 Pro vs GPT-5.2
Gemini 3 Pro and GPT-5.2 score almost the same on the Noometry Index (54.8 vs 54.1), so choose on price, context window or the category you care about most.
Last verified . 59 shared benchmarks.
GoogleSummary
- They share 59 benchmarks with published results for both. Gemini 3 Pro scores higher in 6 categories and GPT-5.2 in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.2 leads 60.0 to 49.9.
- The biggest single-benchmark swing is SimpleBench: 76.4% for Gemini 3 Pro and 45.8% for GPT-5.2.
Side by side
| Gemini 3 Pro | GPT-5.2 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 54.8 | 54.1 |
| Released | 2025-11-18 | 2025-12-11 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $1.75 |
| Output $ / M tokens | — | $14 |
| Results tracked | 67 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 3 Pro: 51.6 (#39), GPT-5.2: 51.6 (#37)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| SWE-bench Verified | 72.9% | 73.8% |
| SWE-bench Verified (bash only) | 74.2% | 72.8% |
| LMArena WebDev | 1440 | 1416 |
| SWE-bench Multilingual | 68.7% | 66.7% |
| GSO | 18.6% | 27.4% |
| WeirdML | 69.9% | 72.2% |
| LMArena Coding | 1481 | 1447 |
| ALE-Bench | 1,177 | 1,294 |
| AlgoTune | 1.83 | 2.05 |
Agentic & Tool Use Too close to call
Gemini 3 Pro: 40.6 (#23), GPT-5.2: 40.2 (#24)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| Terminal-Bench | 69.4% | 64.9% |
| Berkeley Function Calling Leaderboard | 72.5% | 55.9% |
| GDPval | 40.3% | 49.7% |
| Remote Labor Index | 1.3% | 2.5% |
| τ²-bench Airline | 80.5% | 83% |
| τ²-bench Banking | 18% | 32.2% |
| τ²-bench Retail | 75.9% | 81.6% |
| τ²-bench Telecom | 91% | 89.7% |
| DeepResearch Bench | 46.3% | 41.1% |
| LMArena Search | 1207 | 1207 |
| METR Time Horizons | 71% | 75.3% |
| Vending-Bench 2 | 5,478 | 3,591 |
| BALROG | 58.1% | — |
Reasoning Gemini 3 Pro leads
Gemini 3 Pro: 52.5 (#31), GPT-5.2: 50.2 (#35)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| ARC-AGI-2 | 31.1% | 52.9% |
| SimpleBench | 76.4% | 45.8% |
| Kagi LLM Benchmark | 80.1% | 73.3% |
| NYT Connections (extended) | 94.4% | 83.6% |
| ARC-AGI-1 | 75% | 86.2% |
| Chess Puzzles | 31% | 49% |
| EnigmaEval | 18.2% | 10.4% |
| LMArena Hard Prompts | 1480 | 1445 |
| Epoch Capabilities Index | 152.92 | 153.45 |
| ForecastBench | 61.2 | 60.1 |
| CritPt | 6.9% | — |
| EBR-Bench | — | 23% |
| Mystery Game Puzzles | — | 23% |
| DTBench | — | 90.9% |
| LMCA | — | 43.9% |
Math GPT-5.2 leads
Gemini 3 Pro: 49.9 (#59), GPT-5.2: 60.0 (#38)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| MathArena Final-Answer Competitions | 67% | 72% |
| OTIS Mock AIME 2024-2025 | 91.4% | 96.1% |
| ProofBench | 20% | 15% |
| LMArena Math | 1476 | 1440 |
| FrontierMath (Feb 2025 set) | 37.6% | 40.7% |
| FrontierMath Tier 4 (v1) | 18.8% | 18.8% |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 31.7% |
| Omni-MATH | 55.5% | — |
Knowledge Gemini 3 Pro leads
Gemini 3 Pro: 64.4 (#16), GPT-5.2: 59.3 (#32)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 92.6% | 91.4% |
| Humanity's Last Exam | 37.5% | 27.8% |
| Vectara Hallucination Rate | 13.6% | 8.4% |
| LMArena Expert | 1475 | 1445 |
| SimpleQA Verified | — | 37.1% |
| MMLU-Pro | 90.3% | — |
| GPQA (HELM) | 80.3% | — |
Multimodal Gemini 3 Pro leads
Gemini 3 Pro: 57.6 (#2), GPT-5.2: 51.3 (#7)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1305 | 1268 |
| VPCT | 91% | 84% |
| LMArena Document | 1434 | 1405 |
| GeoBench | 84% | — |
| Furniture Assembly | — | 38.3% |
Multilingual Gemini 3 Pro leads
Gemini 3 Pro: 56.9 (#16), GPT-5.2: 53.4 (#67)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1474 | 1425 |
| LMArena Chinese | 1523 | 1460 |
| LMArena French | 1492 | 1455 |
| LMArena German | 1515 | 1448 |
| LMArena Japanese | 1510 | 1420 |
| LMArena Korean | 1448 | 1392 |
| LMArena Russian | 1493 | 1440 |
| LMArena Spanish | 1470 | 1433 |
Instruction Following Gemini 3 Pro leads
Gemini 3 Pro: 76.3 (#45), GPT-5.2: 74.7 (#89)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1458 | 1417 |
| IFEval | 87.7% | — |
Long Context Too close to call
Gemini 3 Pro: 44.0 (#79), GPT-5.2: 44.0 (#78)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| CL-bench | 15.8% | 18.2% |
| LMArena Longer Query | 1471 | 1428 |
Writing & Preference Too close to call
Gemini 3 Pro: 66.4 (#35), GPT-5.2: 66.8 (#32)
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| LMArena Text | 1479 | 1439 |
| LMArena Creative Writing | 1482 | 1401 |
| EQ-Bench Creative Writing | 1525 | 1703 |
| LMArena Multi-Turn | 1484 | 1458 |
| WildBench | 85.9% | — |
Frequently asked questions
Is Gemini 3 Pro better than GPT-5.2?
Gemini 3 Pro and GPT-5.2 score almost the same on the Noometry Index (54.8 vs 54.1), so choose on price, context window or the category you care about most.
Is Gemini 3 Pro or GPT-5.2 better for coding?
They score almost the same on coding (51.6 vs 51.6); test both on your own repository before choosing.
How many benchmarks do Gemini 3 Pro and GPT-5.2 share?
59 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and GPT-5.2 has 67.