# Gemini 2.5 Flash vs Grok 4

> Grok 4 is the stronger model overall, scoring 48.1 to 39.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-2-5-flash-vs-grok-4
- Last updated: 2026-10-11
- Shared benchmarks: 41

## Summary

- They share 41 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 2 categories and Grok 4 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4 leads 36.7 to 18.1.
- The biggest single-benchmark swing is GPQA (HELM): 39% for Gemini 2.5 Flash and 72.7% for Grok 4.

## Snapshot

| | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| Provider | Google | xAI |
| Noometry Index | 39.3 | 48.1 |
| Rank | 170 | 56 |
| Context | 1.05M | — |
| Input $/M | $0.30 | — |
| Output $/M | $2.50 | — |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 2.5 Flash: 35.8 (#220)
- Grok 4: 50.3 (#46)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| Aider Polyglot | 55.1% | 79.6% |
| WeirdML | 41.9% | 45.7% |
| LMArena Coding | 1424 | 1408 |
| SWE-bench Verified (bash only) | 28.7% | — |
| ALE-Bench | 661.88 | — |

## Agentic & Tool Use

- Gemini 2.5 Flash: 30.8 (#74)
- Grok 4: 32.3 (#68)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| Terminal-Bench | 17.1% | 27.2% |
| Berkeley Function Calling Leaderboard | 56.2% | 63% |
| BALROG | 33.5% | 43.6% |
| GDPval | — | 21.1% |
| TheAgentCompany | 41.1% | — |
| Cybench | — | 43% |
| DeepResearch Bench | — | 47.3% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 66.6% |
| Vending-Bench 2 | 548.84 | — |

## Reasoning

- Gemini 2.5 Flash: 18.1 (#286)
- Grok 4: 36.7 (#65)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| ARC-AGI-2 | 2.5% | 16% |
| SimpleBench | 41.2% | 60.5% |
| Kagi LLM Benchmark | 56.8% | 73.6% |
| ARC-AGI-1 | 33.3% | 66.7% |
| LMArena Hard Prompts | 1422 | 1409 |
| Epoch Capabilities Index | 143.03 | 146.44 |
| ForecastBench | 60.6 | 60.9 |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 28% |
| EnigmaEval | 2.7% | — |
| DTBench | 76.5% | — |
| LMCA | 27.5% | — |

## Math

- Gemini 2.5 Flash: 39.9 (#98)
- Grok 4: 48.4 (#64)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 84% |
| Omni-MATH | 38.5% | 60.3% |
| LMArena Math | 1415 | 1422 |
| FrontierMath (Feb 2025 set) | 4.8% | 19.7% |
| FrontierMath Tier 4 (v1) | 4.2% | 2.1% |

## Knowledge

- Gemini 2.5 Flash: 36.4 (#168)
- Grok 4: 53.8 (#55)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| MMLU-Pro | 63.9% | 85.1% |
| Confabulations | 16.8% | 12.4% |
| GPQA (HELM) | 39% | 72.7% |
| LMArena Expert | 1426 | 1415 |
| GPQA Diamond | — | 87% |
| Humanity's Last Exam | 12.1% | — |
| Vectara Hallucination Rate | 7.8% | — |

## Multimodal

- Gemini 2.5 Flash: 41.8 (#32)
- Grok 4: 33.7 (#94)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| LMArena Vision | 1253 | 1210 |
| GeoBench | 76% | 45% |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |

## Multilingual

- Gemini 2.5 Flash: 52.3 (#88)
- Grok 4: 51.8 (#103)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| LMArena Non-English | 1409 | 1403 |
| LMArena Chinese | 1450 | 1427 |
| LMArena French | 1433 | 1418 |
| LMArena German | 1418 | 1429 |
| LMArena Japanese | 1405 | 1394 |
| LMArena Korean | 1385 | 1377 |
| LMArena Russian | 1415 | 1410 |
| LMArena Spanish | 1421 | 1420 |

## Instruction Following

- Gemini 2.5 Flash: 75.7 (#54)
- Grok 4: 79.2 (#5)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| IFEval | 89.8% | 94.9% |
| LMArena Instruction Following | 1405 | 1387 |

## Long Context

- Gemini 2.5 Flash: 47.5 (#17)
- Grok 4: 63.1 (#4)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| Fiction.LiveBench | 77.8% | 94.4% |
| LMArena Longer Query | 1419 | 1409 |

## Writing & Preference

- Gemini 2.5 Flash: 53.8 (#157)
- Grok 4: 58.5 (#116)

| Benchmark | Gemini 2.5 Flash | Grok 4 |
|---|---|---|
| LMArena Text | 1417 | 1411 |
| LMArena Creative Writing | 1400 | 1397 |
| Short-Story Creative Writing | 76.5% | 76.9% |
| WildBench | 81.7% | 79.7% |
| LMArena Multi-Turn | 1408 | 1416 |
| EQ-Bench Creative Writing | 1137 | — |

## FAQ

### Is Gemini 2.5 Flash better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 39.3 on the Noometry Index.

### Is Gemini 2.5 Flash or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 35.8 in the Noometry coding category.

### How many benchmarks do Gemini 2.5 Flash and Grok 4 share?

41 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Grok 4 has 48.