# Grok 4 vs Kimi K2 (Jul 2025)

> Grok 4 is the stronger model overall, scoring 48.1 to 41.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-4-vs-kimi-k2
- Last updated: 2026-10-10
- Shared benchmarks: 36

## Summary

- They share 36 benchmarks with published results for both. Grok 4 scores higher in 7 categories and Kimi K2 (Jul 2025) in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 41.2.
- The biggest single-benchmark swing is SimpleBench: 60.5% for Grok 4 and 26.3% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Provider | xAI | Moonshot AI |
| Noometry Index | 48.1 | 41.2 |
| Rank | 56 | 140 |
| Context | — | 262K |
| Input $/M | — | $0.57 |
| Output $/M | — | $2.30 |
| Weights | Proprietary | Open |

## Coding

- Grok 4: 50.3 (#46)
- Kimi K2 (Jul 2025): 42.4 (#102)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Aider Polyglot | 79.6% | 59.1% |
| WeirdML | 45.7% | 42.8% |
| LMArena Coding | 1408 | 1399 |
| SWE-bench Verified (bash only) | — | 63.4% |
| GSO | — | 4.9% |
| ALE-Bench | — | 597.5 |

## Agentic & Tool Use

- Grok 4: 32.3 (#68)
- Kimi K2 (Jul 2025): 32.4 (#64)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | 27.2% | 35.7% |
| Berkeley Function Calling Leaderboard | 63% | 59.1% |
| METR Time Horizons | 66.6% | 59.2% |
| GDPval | 21.1% | — |
| Cybench | 43% | — |
| DeepResearch Bench | 47.3% | — |
| BALROG | 43.6% | — |
| LMArena Search | 1142 | — |

## Reasoning

- Grok 4: 36.7 (#65)
- Kimi K2 (Jul 2025): 23.3 (#179)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 60.5% | 26.3% |
| Kagi LLM Benchmark | 73.6% | 64.4% |
| LMArena Hard Prompts | 1409 | 1384 |
| Epoch Capabilities Index | 146.44 | 146.01 |
| ForecastBench | 60.9 | 60.2 |
| ARC-AGI-2 | 16% | — |
| ARC-AGI-1 | 66.7% | — |
| Chess Puzzles | 28% | — |

## Math

- Grok 4: 48.4 (#64)
- Kimi K2 (Jul 2025): 42.7 (#83)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 60.3% | 65.4% |
| LMArena Math | 1422 | 1397 |
| FrontierMath (Feb 2025 set) | 19.7% | 21.4% |
| FrontierMath Tier 4 (v1) | 2.1% | 0% |
| OTIS Mock AIME 2024-2025 | 84% | — |

## Knowledge

- Grok 4: 53.8 (#55)
- Kimi K2 (Jul 2025): 37.3 (#157)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 85.1% | 81.9% |
| Confabulations | 12.4% | 20.4% |
| GPQA (HELM) | 72.7% | 65.3% |
| LMArena Expert | 1415 | 1365 |
| GPQA Diamond | 87% | — |
| Vectara Hallucination Rate | — | 17.9% |

## Multimodal

- Grok 4: 33.7 (#94)
- Kimi K2 (Jul 2025): —

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1210 | — |
| GeoBench | 45% | — |

## Multilingual

- Grok 4: 51.8 (#103)
- Kimi K2 (Jul 2025): 49.6 (#130)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1403 | 1372 |
| LMArena Chinese | 1427 | 1415 |
| LMArena French | 1418 | 1379 |
| LMArena German | 1429 | 1387 |
| LMArena Japanese | 1394 | 1349 |
| LMArena Korean | 1377 | 1325 |
| LMArena Russian | 1410 | 1385 |
| LMArena Spanish | 1420 | 1386 |

## Instruction Following

- Grok 4: 79.2 (#5)
- Kimi K2 (Jul 2025): 71.1 (#156)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 94.9% | 85% |
| LMArena Instruction Following | 1387 | 1348 |

## Long Context

- Grok 4: 63.1 (#4)
- Kimi K2 (Jul 2025): 41.2 (#145)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 94.4% | 66.7% |
| LMArena Longer Query | 1409 | 1353 |
| CL-bench | — | 17.6% |

## Writing & Preference

- Grok 4: 58.5 (#116)
- Kimi K2 (Jul 2025): 62.3 (#78)

| Benchmark | Grok 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1411 | 1380 |
| LMArena Creative Writing | 1397 | 1350 |
| Short-Story Creative Writing | 76.9% | 85.6% |
| WildBench | 79.7% | 86.2% |
| LMArena Multi-Turn | 1416 | 1371 |
| EQ-Bench Creative Writing | — | 1666 |

## FAQ

### Is Grok 4 better than Kimi K2 (Jul 2025)?

Grok 4 is the stronger model overall, scoring 48.1 to 41.2 on the Noometry Index.

### Is Grok 4 or Kimi K2 (Jul 2025) better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 42.4 in the Noometry coding category.

### How many benchmarks do Grok 4 and Kimi K2 (Jul 2025) share?

36 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Kimi K2 (Jul 2025) has 42.