# DeepSeek LLM 67B vs Kimi K2.6

> Kimi K2.6 is the stronger model overall, scoring 47.5 to 24.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-kimi-k2-6
- Last updated: 2026-10-10
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Kimi K2.6 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 56.8 to 8.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 96.1% for Kimi K2.6.

## Snapshot

| | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 24.8 | 47.5 |
| Rank | 341 | 52 |
| Context | — | 262K |
| Input $/M | — | $0.95 |
| Output $/M | — | $4 |
| Weights | Open | Open |

## Coding

- DeepSeek LLM 67B: 31.3 (#273)
- Kimi K2.6: 50.5 (#41)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| LMArena Coding | 1096 | 1488 |
| SWE-bench Verified | — | 76.7% |
| LMArena WebDev | — | 1509 |
| SciCode | — | 53.5% |
| WeirdML | — | 55.9% |
| ALE-Bench | — | 1,093 |

## Agentic & Tool Use

- DeepSeek LLM 67B: —
- Kimi K2.6: 21.8 (#128)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| OSWorld 2.0 | — | 4.6% |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| Vending-Bench 2 | — | 6,205 |

## Reasoning

- DeepSeek LLM 67B: 15.9 (#300)
- Kimi K2.6: 39.6 (#51)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| Chess Puzzles | 0% | 26% |
| LMArena Hard Prompts | 1070 | 1470 |
| Epoch Capabilities Index | 110.54 | 151.03 |
| NYT Connections (extended) | — | 87.2% |
| CritPt | — | 8% |
| EBR-Bench | — | 2.4% |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 90.9% |
| LMCA | — | 37.3% |

## Math

- DeepSeek LLM 67B: 8.1 (#318)
- Kimi K2.6: 56.8 (#34)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 96.1% |
| LMArena Math | 1108 | 1475 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| ProofBench | — | 16% |
| MATH Level 5 | 6.4% | — |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |

## Knowledge

- DeepSeek LLM 67B: 7.1 (#308)
- Kimi K2.6: 53.9 (#48)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 24.6% | 90.8% |
| SimpleQA Verified | — | 34.9% |
| Vectara Hallucination Rate | — | 10.8% |
| LMArena Expert | — | 1491 |

## Multimodal

- DeepSeek LLM 67B: —
- Kimi K2.6: 31.8 (#95)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| LMArena Vision | — | 1282 |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |

## Multilingual

- DeepSeek LLM 67B: 29.5 (#262)
- Kimi K2.6: 55.1 (#35)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1073 | 1446 |
| LMArena Chinese | 1132 | 1521 |
| LMArena French | — | 1471 |
| LMArena German | — | 1449 |
| LMArena Japanese | — | 1442 |
| LMArena Korean | — | 1428 |
| LMArena Russian | — | 1446 |
| LMArena Spanish | — | 1463 |

## Instruction Following

- DeepSeek LLM 67B: 55.7 (#272)
- Kimi K2.6: 76.5 (#40)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1079 | 1451 |

## Long Context

- DeepSeek LLM 67B: 33.1 (#260)
- Kimi K2.6: 44.9 (#50)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1092 | 1468 |

## Writing & Preference

- DeepSeek LLM 67B: 31.8 (#277)
- Kimi K2.6: 68.7 (#25)

| Benchmark | DeepSeek LLM 67B | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1105 | 1455 |
| LMArena Creative Writing | 1067 | 1434 |
| LMArena Multi-Turn | 1082 | 1453 |
| EQ-Bench Creative Writing | — | 1725 |
| EQ-Bench 4 | — | 1202 |

## FAQ

### Is DeepSeek LLM 67B better than Kimi K2.6?

Kimi K2.6 is the stronger model overall, scoring 47.5 to 24.8 on the Noometry Index.

### Is DeepSeek LLM 67B or Kimi K2.6 better for coding?

Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 31.3 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and Kimi K2.6 share?

14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Kimi K2.6 has 51.