Model comparison
Gemini 2.5 Pro vs Grok 4.6
Grok 4.6 is the stronger model overall, scoring 56.9 to 45.0 on the Noometry Index.
Last verified . 36 shared benchmarks.
GoogleSummary
- They share 36 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 4 categories and Grok 4.6 in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.6 leads 67.0 to 32.5.
- The biggest single-benchmark swing is ARC-AGI-2: 4.9% for Gemini 2.5 Pro and 67.1% for Grok 4.6.
- Grok 4.6 is cheaper at $2 / $6 per million input/output tokens, against $1.25 / $10 for Gemini 2.5 Pro.
- Gemini 2.5 Pro accepts more context: 1.05M tokens versus 500K.
Side by side
| Gemini 2.5 Pro | Grok 4.6 | |
|---|---|---|
| Provider | xAI | |
| Noometry Index | 45.0 | 56.9 |
| Released | 2025-03-25 | 2026-08-12 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 500K |
| Max output | 66K | 500K |
| Input $ / M tokens | $1.25 | $2 |
| Output $ / M tokens | $10 | $6 |
| Results tracked | 78 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
Gemini 2.5 Pro: 42.4 (#101), Grok 4.6: 58.5 (#16)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| LMArena WebDev | 1227 | 1617 |
| SciCode | 42.8% | 56.5% |
| WeirdML | 54% | 67.3% |
| LMArena Coding | 1452 | 1465 |
| ALE-Bench | 785.52 | 1,508 |
| SWE-bench Verified | 57.6% | — |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| SWE-bench Verified (bash only) | 53.6% | — |
| Aider Polyglot | 83.1% | — |
| CursorBench | — | 41.4% |
| FrontierSWE | — | 25.3% |
| GSO | 3.9% | — |
| LiveBench Coding | 85.9% | — |
| CadEval | 64% | — |
| AlgoTune | 1.51 | — |
Agentic & Tool Use Grok 4.6 leads
Gemini 2.5 Pro: 29.2 (#88), Grok 4.6: 39.4 (#27)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| Vending-Bench 2 | 573.64 | 9,047 |
| Terminal-Bench | 32.6% | — |
| APEX-Agents | — | 65.3% |
| GDPval | 23.3% | — |
| Remote Labor Index | 0.8% | — |
| TheAgentCompany | 30.3% | — |
| τ²-bench Banking | 13.7% | — |
| DeepResearch Bench | 42.8% | — |
| BALROG | 43.3% | — |
| GDP.pdf | — | 17.2% |
| LMArena Search | 1142 | — |
| METR Time Horizons | 55.4% | — |
Reasoning Grok 4.6 leads
Gemini 2.5 Pro: 28.8 (#99), Grok 4.6: 61.4 (#20)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 4.9% | 67.1% |
| SimpleBench | 62.4% | 75.9% |
| ARC-AGI-1 | 41% | 87.5% |
| CritPt | 2% | 19.7% |
| Chess Puzzles | 20% | 40% |
| LMArena Hard Prompts | 1455 | 1447 |
| DTBench | 82.4% | 97.3% |
| LMCA | 34.8% | 48.5% |
| Epoch Capabilities Index | 145.32 | 156.44 |
| Kagi LLM Benchmark | 70.3% | — |
| NYT Connections (extended) | — | 80% |
| EnigmaEval | 5.6% | — |
| EBR-Bench | — | 30.5% |
| LiveBench Reasoning | 89.8% | — |
| Mystery Game Puzzles | — | 34% |
| LiveBench Data Analysis | 79.9% | — |
| ForecastBench | 61.3 | — |
| LiveBench | 82.3% | — |
Math Grok 4.6 leads
Gemini 2.5 Pro: 32.5 (#213), Grok 4.6: 67.0 (#24)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 24.6% | 66% |
| FrontierMath Tier 4 | 0% | 31.7% |
| OTIS Mock AIME 2024-2025 | 84.7% | 99.2% |
| LMArena Math | 1450 | 1423 |
| ProofBench | — | 51% |
| Omni-MATH | 41.6% | — |
| LiveBench Math | 90.2% | — |
| MATH Level 5 | 95.9% | — |
| FrontierMath (Feb 2025 set) | 14.1% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Grok 4.6 leads
Gemini 2.5 Pro: 56.0 (#46), Grok 4.6: 63.3 (#20)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 85.3% | 94% |
| LMArena Expert | 1452 | 1467 |
| Humanity's Last Exam | 21.6% | — |
| SimpleQA Verified | — | 49.3% |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.6% | — |
| Vectara Hallucination Rate | 7% | — |
| GPQA (HELM) | 74.9% | — |
Multimodal Gemini 2.5 Pro leads
Gemini 2.5 Pro: 45.2 (#18), Grok 4.6: 43.6 (#23)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| LMArena Vision | 1263 | 1263 |
| LMArena Document | 1421 | 1452 |
| GeoBench | 86% | — |
| VPCT | 48% | — |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| SpatialViz-Bench | 44.7% | — |
Multilingual Gemini 2.5 Pro leads
Gemini 2.5 Pro: 55.3 (#31), Grok 4.6: 53.0 (#74)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1451 | 1420 |
| LMArena Chinese | 1507 | 1480 |
| LMArena French | 1472 | 1461 |
| LMArena German | 1487 | 1431 |
| LMArena Japanese | 1461 | 1376 |
| LMArena Korean | 1434 | 1397 |
| LMArena Russian | 1461 | 1422 |
| LMArena Spanish | 1473 | 1404 |
Instruction Following Too close to call
Gemini 2.5 Pro: 75.0 (#75), Grok 4.6: 75.4 (#63)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1437 | 1431 |
| LiveBench Instruction Following | 80.6% | — |
| IFEval | 84% | — |
Long Context Gemini 2.5 Pro leads
Gemini 2.5 Pro: 59.8 (#5), Grok 4.6: 44.5 (#66)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1449 | 1454 |
| Fiction.LiveBench | 91.7% | — |
Writing & Preference Gemini 2.5 Pro leads
Gemini 2.5 Pro: 63.7 (#62), Grok 4.6: 62.3 (#80)
| Benchmark | Gemini 2.5 Pro | Grok 4.6 |
|---|---|---|
| LMArena Text | 1458 | 1428 |
| LMArena Creative Writing | 1454 | 1428 |
| LMArena Multi-Turn | 1453 | 1425 |
| Short-Story Creative Writing | 83.8% | — |
| EQ-Bench Creative Writing | 1421 | — |
| WildBench | 85.7% | — |
| LiveBench Language | 67.8% | — |
Frequently asked questions
Is Gemini 2.5 Pro better than Grok 4.6?
Grok 4.6 is the stronger model overall, scoring 56.9 to 45.0 on the Noometry Index.
Which is cheaper, Gemini 2.5 Pro or Grok 4.6?
Grok 4.6 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.
Is Gemini 2.5 Pro or Grok 4.6 better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Pro does, with 1.05M tokens against 500K.
How many benchmarks do Gemini 2.5 Pro and Grok 4.6 share?
36 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and Grok 4.6 has 49.