# Grok-2 (Dec 2024) vs Kimi K2.5 Instant

> Kimi K2.5 Instant is the stronger model overall, scoring 42.9 to 33.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-2-vs-kimi-k2-5-instant
- Last updated: 2026-10-10
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Grok-2 (Dec 2024) scores higher in 0 categories and Kimi K2.5 Instant in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.5 Instant leads 36.6 to 19.6.
- Kimi K2.5 Instant has downloadable open weights; the other is API-only.

## Snapshot

| | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| Provider | xAI | Moonshot AI |
| Noometry Index | 33.4 | 42.9 |
| Rank | 233 | 92 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Open |

## Coding

- Grok-2 (Dec 2024): 32.9 (#254)
- Kimi K2.5 Instant: 42.3 (#95)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Coding | 1288 | 1483 |
| LMArena WebDev | — | 1404 |
| WeirdML | 22.2% | — |
| LiveBench Coding | 46.4% | — |

## Reasoning

- Grok-2 (Dec 2024): 16.4 (#294)
- Kimi K2.5 Instant: 28.5 (#88)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Hard Prompts | 1272 | 1443 |
| SimpleBench | 22.7% | — |
| LiveBench Reasoning | 54.8% | — |
| DTBench | 65.2% | — |
| LiveBench Data Analysis | 54.5% | — |
| Epoch Capabilities Index | 130.48 | — |
| LiveBench | 54.3% | — |

## Math

- Grok-2 (Dec 2024): 19.6 (#278)
- Kimi K2.5 Instant: 36.6 (#105)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Math | 1283 | 1443 |
| OTIS Mock AIME 2024-2025 | 11.5% | — |
| LiveBench Math | 54.9% | — |
| MATH Level 5 | 63.5% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |

## Knowledge

- Grok-2 (Dec 2024): 29.8 (#225)
- Kimi K2.5 Instant: 39.5 (#121)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Expert | 1254 | 1438 |
| GPQA Diamond | 53.8% | — |
| Confabulations | 20.1% | — |

## Multimodal

- Grok-2 (Dec 2024): —
- Kimi K2.5 Instant: 39.5 (#47)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Vision | — | 1254 |

## Multilingual

- Grok-2 (Dec 2024): 43.3 (#183)
- Kimi K2.5 Instant: 52.2 (#89)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Non-English | 1282 | 1406 |
| LMArena Chinese | 1289 | 1450 |
| LMArena French | 1317 | 1403 |
| LMArena German | 1287 | 1413 |
| LMArena Korean | 1237 | 1378 |
| LMArena Russian | 1286 | 1404 |
| LMArena Spanish | 1281 | 1448 |
| LMArena Japanese | 1244 | — |

## Instruction Following

- Grok-2 (Dec 2024): 67.1 (#197)
- Kimi K2.5 Instant: 75.5 (#61)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Instruction Following | 1270 | 1430 |
| LiveBench Instruction Following | 69.6% | — |

## Long Context

- Grok-2 (Dec 2024): 38.7 (#186)
- Kimi K2.5 Instant: 43.8 (#82)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Longer Query | 1276 | 1435 |

## Writing & Preference

- Grok-2 (Dec 2024): 48.8 (#192)
- Kimi K2.5 Instant: 60.8 (#91)

| Benchmark | Grok-2 (Dec 2024) | Kimi K2.5 Instant |
|---|---|---|
| LMArena Text | 1305 | 1420 |
| LMArena Creative Writing | 1284 | 1379 |
| LMArena Multi-Turn | 1290 | 1427 |
| Short-Story Creative Writing | 63.6% | — |
| LiveBench Language | 45.6% | — |

## FAQ

### Is Grok-2 (Dec 2024) better than Kimi K2.5 Instant?

Kimi K2.5 Instant is the stronger model overall, scoring 42.9 to 33.4 on the Noometry Index.

### Is Grok-2 (Dec 2024) or Kimi K2.5 Instant better for coding?

Kimi K2.5 Instant scores higher on coding benchmarks: 42.3 versus 32.9 in the Noometry coding category.

### How many benchmarks do Grok-2 (Dec 2024) and Kimi K2.5 Instant share?

16 benchmarks have published results for both models. Grok-2 (Dec 2024) has 34 scored results on Noometry and Kimi K2.5 Instant has 18.