# Deepseek Coder v2 vs Kimi K2 (Jul 2025)

> Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 35.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-coder-v2-vs-kimi-k2
- Last updated: 2026-10-12
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and Kimi K2 (Jul 2025) in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.3 to 38.2.

## Snapshot

| | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 35.9 | 41.2 |
| Rank | 220 | 140 |
| Context | — | 262K |
| Input $/M | — | $0.57 |
| Output $/M | — | $2.30 |
| Weights | Open | Open |

## Coding

- Deepseek Coder v2: 38.1 (#183)
- Kimi K2 (Jul 2025): 42.4 (#102)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Coding | 1251 | 1399 |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| GSO | — | 4.9% |
| WeirdML | — | 42.8% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| ALE-Bench | — | 597.5 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |

## Agentic & Tool Use

- Deepseek Coder v2: —
- Kimi K2 (Jul 2025): 32.4 (#64)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| METR Time Horizons | — | 59.2% |

## Reasoning

- Deepseek Coder v2: 23.6 (#176)
- Kimi K2 (Jul 2025): 23.3 (#179)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1384 |
| SimpleBench | — | 26.3% |
| Kagi LLM Benchmark | — | 64.4% |
| Epoch Capabilities Index | — | 146.01 |
| ForecastBench | — | 60.2 |
| WinoGrande | 83.7% | — |

## Math

- Deepseek Coder v2: 34.9 (#190)
- Kimi K2 (Jul 2025): 42.7 (#83)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1241 | 1397 |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 94.5% | — |

## Knowledge

- Deepseek Coder v2: 32.3 (#212)
- Kimi K2 (Jul 2025): 37.3 (#157)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1181 | 1365 |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
| ARC (AI2) Challenge | 64.3% | — |

## Multilingual

- Deepseek Coder v2: 36.3 (#240)
- Kimi K2 (Jul 2025): 49.6 (#130)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1182 | 1372 |
| LMArena Chinese | 1201 | 1415 |
| LMArena French | 1185 | 1379 |
| LMArena German | 1164 | 1387 |
| LMArena Japanese | 1126 | 1349 |
| LMArena Korean | 1104 | 1325 |
| LMArena Russian | 1188 | 1385 |
| LMArena Spanish | 1153 | 1386 |

## Instruction Following

- Deepseek Coder v2: 61.7 (#242)
- Kimi K2 (Jul 2025): 71.1 (#156)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1180 | 1348 |
| IFEval | — | 85% |

## Long Context

- Deepseek Coder v2: 37.0 (#224)
- Kimi K2 (Jul 2025): 41.2 (#145)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1219 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |

## Writing & Preference

- Deepseek Coder v2: 38.2 (#253)
- Kimi K2 (Jul 2025): 62.3 (#78)

| Benchmark | Deepseek Coder v2 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1191 | 1380 |
| LMArena Creative Writing | 1120 | 1350 |
| LMArena Multi-Turn | 1177 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
| WildBench | — | 86.2% |

## FAQ

### Is Deepseek Coder v2 better than Kimi K2 (Jul 2025)?

Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 35.9 on the Noometry Index.

### Is Deepseek Coder v2 or Kimi K2 (Jul 2025) better for coding?

Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 38.1 in the Noometry coding category.

### How many benchmarks do Deepseek Coder v2 and Kimi K2 (Jul 2025) share?

17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Kimi K2 (Jul 2025) has 42.