Model comparison
Gemini 3.1 Pro Preview vs Grok 4
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 48.1 on the Noometry Index.
Last verified . 34 shared benchmarks.
GoogleSummary
- They share 34 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 7 categories and Grok 4 in 3 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 36.7.
- The biggest single-benchmark swing is ARC-AGI-2: 77.1% for Gemini 3.1 Pro Preview and 16% for Grok 4.
Side by side
| Gemini 3.1 Pro Preview | Grok 4 | |
|---|---|---|
| Provider | xAI | |
| Noometry Index | 56.7 | 48.1 |
| Released | 2026-02-19 | 2025-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $12 | — |
| Results tracked | 71 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4 leads
Gemini 3.1 Pro Preview: 42.5 (#99), Grok 4: 50.3 (#46)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| WeirdML | 72.1% | 45.7% |
| LMArena Coding | 1484 | 1408 |
| SWE-bench Verified | 75.6% | — |
| DeepSWE | 11.7% | — |
| Aider Polyglot | — | 79.6% |
| LMArena WebDev | 1447 | — |
| SciCode | 58.9% | — |
| GSO | 22.6% | — |
| MirrorCode | 8.9% | — |
| ALE-Bench | 1,161 | — |
| AlgoTune | 2.02 | — |
Agentic & Tool Use Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 37.7 (#34), Grok 4: 32.3 (#68)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| Terminal-Bench | 80.2% | 27.2% |
| DeepResearch Bench | 47.8% | 47.3% |
| BALROG | 57% | 43.6% |
| LMArena Search | 1211 | 1142 |
| METR Time Horizons | 77% | 66.6% |
| APEX-Agents | 35.3% | — |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 21.1% |
| τ²-bench Banking | 26% | — |
| Cybench | — | 43% |
| PostTrainBench | 22% | — |
| ExploitBench | 26.1% | — |
| GBAEval | 0.8% | — |
| GDP.pdf | 17% | — |
| Vending-Bench 2 | 3,774 | — |
Reasoning Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 71.7 (#12), Grok 4: 36.7 (#65)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| ARC-AGI-2 | 77.1% | 16% |
| SimpleBench | 79.6% | 60.5% |
| ARC-AGI-1 | 98% | 66.7% |
| Chess Puzzles | 55% | 28% |
| LMArena Hard Prompts | 1485 | 1409 |
| Epoch Capabilities Index | 154.77 | 146.44 |
| ForecastBench | 59 | 60.9 |
| Kagi LLM Benchmark | — | 73.6% |
| NYT Connections (extended) | 97.4% | — |
| CritPt | 17.7% | — |
| EnigmaEval | 36.8% | — |
| Thematic Generalization | 79.4% | — |
| EBR-Bench | 14.3% | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 97.1% | — |
| LMCA | 53.8% | — |
Math Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 62.1 (#34), Grok 4: 48.4 (#64)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 95.6% | 84% |
| LMArena Math | 1485 | 1422 |
| FrontierMath (Feb 2025 set) | 36.9% | 19.7% |
| FrontierMath Tier 4 (v1) | 16.7% | 2.1% |
| FrontierMath (Tiers 1-3) | 59.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 86.5% | — |
| ProofBench | 26% | — |
| Omni-MATH | — | 60.3% |
Knowledge Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 71.8 (#3), Grok 4: 53.8 (#55)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| GPQA Diamond | 94.4% | 87% |
| LMArena Expert | 1485 | 1415 |
| Humanity's Last Exam | 46.4% | — |
| SimpleQA Verified | 73.5% | — |
| MMLU-Pro | — | 85.1% |
| Confabulations | — | 12.4% |
| Vectara Hallucination Rate | 10.4% | — |
| GPQA (HELM) | — | 72.7% |
Multimodal Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 37.9 (#69), Grok 4: 33.7 (#94)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| LMArena Vision | 1296 | 1210 |
| GeoBench | — | 45% |
| Blueprint-Bench 2 | 26.5% | — |
| Furniture Assembly | 26.7% | — |
| LMArena Document | 1444 | — |
Multilingual Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 57.0 (#12), Grok 4: 51.8 (#103)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| LMArena Non-English | 1477 | 1403 |
| LMArena Chinese | 1529 | 1427 |
| LMArena French | 1487 | 1418 |
| LMArena German | 1491 | 1429 |
| LMArena Japanese | 1493 | 1394 |
| LMArena Korean | 1455 | 1377 |
| LMArena Russian | 1498 | 1410 |
| LMArena Spanish | 1479 | 1420 |
Instruction Following Grok 4 leads
Gemini 3.1 Pro Preview: 77.0 (#32), Grok 4: 79.2 (#5)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| LMArena Instruction Following | 1466 | 1387 |
| IFEval | — | 94.9% |
Long Context Grok 4 leads
Gemini 3.1 Pro Preview: 47.4 (#18), Grok 4: 63.1 (#4)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1483 | 1409 |
| Fiction.LiveBench | — | 94.4% |
| CL-bench | 20.8% | — |
| CL-bench Life | 16.9% | — |
Writing & Preference Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 66.1 (#37), Grok 4: 58.5 (#116)
| Benchmark | Gemini 3.1 Pro Preview | Grok 4 |
|---|---|---|
| LMArena Text | 1481 | 1411 |
| LMArena Creative Writing | 1482 | 1397 |
| LMArena Multi-Turn | 1488 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| EQ-Bench Creative Writing | 1491 | — |
| WildBench | — | 79.7% |
| EQ-Bench 4 | 1142 | — |
Frequently asked questions
Is Gemini 3.1 Pro Preview better than Grok 4?
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 48.1 on the Noometry Index.
Is Gemini 3.1 Pro Preview or Grok 4 better for coding?
Grok 4 scores higher on coding benchmarks: 50.3 versus 42.5 in the Noometry coding category.
How many benchmarks do Gemini 3.1 Pro Preview and Grok 4 share?
34 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Grok 4 has 48.