# Kimi K2.6 vs Kimi K2.7 Code

> Kimi K2.6 is the stronger model overall, scoring 47.5 to 43.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/kimi-k2-6-vs-kimi-k2-7-code
- Last updated: 2026-10-10
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. Kimi K2.6 scores higher in 4 categories and Kimi K2.7 Code in 1 category; 4 gaps are clear of the uncertainty.
- The widest gap is in coding, where Kimi K2.6 leads 50.5 to 42.4.
- The biggest single-benchmark swing is FrontierMath Tier 4: 25.6% for Kimi K2.6 and 12.2% for Kimi K2.7 Code.
- Both cost about the same: $0.95 input and $4 output per million tokens.

## Snapshot

| | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| Provider | Moonshot AI | Moonshot AI |
| Noometry Index | 47.5 | 43.0 |
| Rank | 52 | 90 |
| Context | 262K | 262K |
| Input $/M | $0.95 | $0.95 |
| Output $/M | $4 | $4 |
| Weights | Open | Open |

## Coding

- Kimi K2.6: 50.5 (#41)
- Kimi K2.7 Code: 42.4 (#92)

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| LMArena WebDev | 1509 | 1473 |
| SciCode | 53.5% | 47.5% |
| WeirdML | 55.9% | 54.1% |
| ALE-Bench | 1,093 | 886.23 |
| SWE-bench Verified | 76.7% | — |
| DeepSWE | — | 30.5% |
| FrontierCode | — | 30.1% |
| LMArena Coding | 1488 | — |

## Agentic & Tool Use

- Kimi K2.6: 21.8 (#128)
- Kimi K2.7 Code: 23.9 (#115)

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| GBAEval | 0.9% | 0.9% |
| Vending-Bench 2 | 6,205 | 5,083 |
| APEX-Agents | — | 37.6% |
| OSWorld 2.0 | 4.6% | — |
| ExploitBench | 18.4% | — |
| GDP.pdf | 12% | — |

## Reasoning

- Kimi K2.6: 39.6 (#51)
- Kimi K2.7 Code: 38.2 (#56)

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| CritPt | 8% | 10% |
| Chess Puzzles | 26% | 21% |
| Epoch Capabilities Index | 151.03 | 150.01 |
| SimpleBench | — | 57.9% |
| NYT Connections (extended) | 87.2% | — |
| EBR-Bench | 2.4% | — |
| LMArena Hard Prompts | 1470 | — |
| Mystery Game Puzzles | 18% | — |
| DTBench | 90.9% | — |
| LMCA | 37.3% | — |
| Surface Evolver Bench | — | 48.8% |

## Math

- Kimi K2.6: 56.8 (#34)
- Kimi K2.7 Code: 52.6 (#39)

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| FrontierMath (Tiers 1-3) | 57.2% | 54% |
| FrontierMath Tier 4 | 25.6% | 12.2% |
| OTIS Mock AIME 2024-2025 | 96.1% | 95.6% |
| MathArena Final-Answer Competitions | 72.9% | — |
| ProofBench | 16% | — |
| LMArena Math | 1475 | — |
| FrontierMath (Feb 2025 set) | 39% | — |
| FrontierMath Tier 4 (v1) | 14.6% | — |

## Knowledge

- Kimi K2.6: 53.9 (#48)
- Kimi K2.7 Code: 53.6 (#51)

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| GPQA Diamond | 90.8% | 87.9% |
| SimpleQA Verified | 34.9% | 36.5% |
| Vectara Hallucination Rate | 10.8% | — |
| LMArena Expert | 1491 | — |

## Multimodal

- Kimi K2.6: 31.8 (#95)
- Kimi K2.7 Code: —

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| LMArena Vision | 1282 | — |
| Blueprint-Bench 2 | 3.9% | — |
| Furniture Assembly | 21.7% | — |
| LMArena Document | 1451 | — |

## Multilingual

- Kimi K2.6: 55.1 (#35)
- Kimi K2.7 Code: —

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| LMArena Non-English | 1446 | — |
| LMArena Chinese | 1521 | — |
| LMArena French | 1471 | — |
| LMArena German | 1449 | — |
| LMArena Japanese | 1442 | — |
| LMArena Korean | 1428 | — |
| LMArena Russian | 1446 | — |
| LMArena Spanish | 1463 | — |

## Instruction Following

- Kimi K2.6: 76.5 (#40)
- Kimi K2.7 Code: —

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| LMArena Instruction Following | 1451 | — |

## Long Context

- Kimi K2.6: 44.9 (#50)
- Kimi K2.7 Code: —

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| LMArena Longer Query | 1468 | — |

## Writing & Preference

- Kimi K2.6: 68.7 (#25)
- Kimi K2.7 Code: —

| Benchmark | Kimi K2.6 | Kimi K2.7 Code |
|---|---|---|
| LMArena Text | 1455 | — |
| LMArena Creative Writing | 1434 | — |
| EQ-Bench Creative Writing | 1725 | — |
| EQ-Bench 4 | 1202 | — |
| LMArena Multi-Turn | 1453 | — |

## FAQ

### Is Kimi K2.6 better than Kimi K2.7 Code?

Kimi K2.6 is the stronger model overall, scoring 47.5 to 43.0 on the Noometry Index.

### Which is cheaper, Kimi K2.6 or Kimi K2.7 Code?

Kimi K2.7 Code is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; Kimi K2.6 lists at $0.95 and $4.

### Is Kimi K2.6 or Kimi K2.7 Code better for coding?

Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 42.4 in the Noometry coding category.

### Which has the bigger context window?

Both accept 262K tokens.

### How many benchmarks do Kimi K2.6 and Kimi K2.7 Code share?

14 benchmarks have published results for both models. Kimi K2.6 has 51 scored results on Noometry and Kimi K2.7 Code has 19.