Model comparison
Gemini 3 Pro vs Grok 3
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 39.9 on the Noometry Index.
Last verified . 35 shared benchmarks.
GoogleSummary
- They share 35 benchmarks with published results for both. Gemini 3 Pro scores higher in 9 categories and Grok 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 13.7.
- The biggest single-benchmark swing is ARC-AGI-1: 75% for Gemini 3 Pro and 5.5% for Grok 3.
Side by side
| Gemini 3 Pro | Grok 3 | |
|---|---|---|
| Provider | xAI | |
| Noometry Index | 54.8 | 39.9 |
| Released | 2025-11-18 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 67 | 40 |
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), Grok 3: 41.9 (#115)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| WeirdML | 69.9% | 37.2% |
| LMArena Coding | 1481 | 1432 |
| SWE-bench Verified | 72.9% | — |
| SWE-bench Verified (bash only) | 74.2% | — |
| Aider Polyglot | — | 53.3% |
| LMArena WebDev | 1440 | — |
| SWE-bench Multilingual | 68.7% | — |
| GSO | 18.6% | — |
| ALE-Bench | 1,177 | — |
| AlgoTune | 1.83 | — |
Agentic & Tool Use Gemini 3 Pro leads
Gemini 3 Pro: 40.6 (#23), Grok 3: 30.5 (#76)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| BALROG | 58.1% | 29.5% |
| Terminal-Bench | 69.4% | — |
| Berkeley Function Calling Leaderboard | 72.5% | — |
| 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% | — |
| LMArena Search | 1207 | — |
| METR Time Horizons | 71% | — |
| Vending-Bench 2 | 5,478 | — |
Reasoning Gemini 3 Pro leads
Gemini 3 Pro: 52.5 (#31), Grok 3: 13.7 (#333)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| ARC-AGI-2 | 31.1% | 0% |
| SimpleBench | 76.4% | 36.1% |
| Kagi LLM Benchmark | 80.1% | 61.3% |
| ARC-AGI-1 | 75% | 5.5% |
| LMArena Hard Prompts | 1480 | 1434 |
| Epoch Capabilities Index | 152.92 | 138.33 |
| NYT Connections (extended) | 94.4% | — |
| CritPt | 6.9% | — |
| Chess Puzzles | 31% | — |
| EnigmaEval | 18.2% | — |
| ForecastBench | 61.2 | — |
Math Gemini 3 Pro leads
Gemini 3 Pro: 49.9 (#59), Grok 3: 38.0 (#145)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.4% | 55.6% |
| Omni-MATH | 55.5% | 46.4% |
| LMArena Math | 1476 | 1391 |
| FrontierMath (Feb 2025 set) | 37.6% | 3.8% |
| FrontierMath Tier 4 (v1) | 18.8% | 0% |
| MathArena Final-Answer Competitions | 67% | — |
| ProofBench | 20% | — |
| MATH Level 5 | — | 88.7% |
Knowledge Gemini 3 Pro leads
Gemini 3 Pro: 64.4 (#16), Grok 3: 46.2 (#82)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| GPQA Diamond | 92.6% | 75.8% |
| MMLU-Pro | 90.3% | 78.8% |
| Vectara Hallucination Rate | 13.6% | 5.8% |
| GPQA (HELM) | 80.3% | 65% |
| LMArena Expert | 1475 | 1421 |
| Humanity's Last Exam | 37.5% | — |
| Confabulations | — | 14.2% |
Multimodal Not comparable
Gemini 3 Pro: 57.6 (#2), Grok 3: —
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| LMArena Vision | 1305 | — |
| GeoBench | 84% | — |
| VPCT | 91% | — |
| LMArena Document | 1434 | — |
Multilingual Gemini 3 Pro leads
Gemini 3 Pro: 56.9 (#16), Grok 3: 52.3 (#87)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| LMArena Non-English | 1474 | 1410 |
| LMArena Chinese | 1523 | 1448 |
| LMArena French | 1492 | 1460 |
| LMArena German | 1515 | 1431 |
| LMArena Japanese | 1510 | 1387 |
| LMArena Korean | 1448 | 1373 |
| LMArena Russian | 1493 | 1416 |
| LMArena Spanish | 1470 | 1417 |
Instruction Following Gemini 3 Pro leads
Gemini 3 Pro: 76.3 (#45), Grok 3: 75.0 (#73)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| IFEval | 87.7% | 88.4% |
| LMArena Instruction Following | 1458 | 1409 |
Long Context Gemini 3 Pro leads
Gemini 3 Pro: 44.0 (#79), Grok 3: 38.7 (#192)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1471 | 1439 |
| Fiction.LiveBench | — | 58.3% |
| CL-bench | 15.8% | — |
Writing & Preference Gemini 3 Pro leads
Gemini 3 Pro: 66.4 (#35), Grok 3: 55.8 (#141)
| Benchmark | Gemini 3 Pro | Grok 3 |
|---|---|---|
| LMArena Text | 1479 | 1426 |
| LMArena Creative Writing | 1482 | 1414 |
| EQ-Bench Creative Writing | 1525 | 1186 |
| WildBench | 85.9% | 84.9% |
| LMArena Multi-Turn | 1484 | 1425 |
| Short-Story Creative Writing | — | 76.4% |
Frequently asked questions
Is Gemini 3 Pro better than Grok 3?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 39.9 on the Noometry Index.
Is Gemini 3 Pro or Grok 3 better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 51.6 versus 41.9 in the Noometry coding category.
How many benchmarks do Gemini 3 Pro and Grok 3 share?
35 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and Grok 3 has 40.