# Kimi K2.7 Code vs Qwen2.5 7B Instruct

> Kimi K2.7 Code is the stronger model overall, scoring 43.3 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 5.6× less per token, which makes it the better buy when Kimi K2.7 Code's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/kimi-k2-7-code-vs-qwen2-5-7b-instruct
- Last updated: 2026-10-12
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. Kimi K2.7 Code scores higher in 5 categories and Qwen2.5 7B Instruct in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.7 Code leads 52.9 to 12.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 95.6% for Kimi K2.7 Code and 2.5% for Qwen2.5 7B Instruct.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $0.95 / $4 for Kimi K2.7 Code.
- Kimi K2.7 Code accepts more context: 262K tokens versus 131K.

## Snapshot

| | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| Provider | Moonshot AI | Alibaba (Qwen) |
| Noometry Index | 43.3 | 29.0 |
| Rank | 94 | 320 |
| Context | 262K | 131K |
| Input $/M | $0.95 | $0.17 |
| Output $/M | $4 | $0.70 |
| Weights | Open | Open |

## Coding

- Kimi K2.7 Code: 42.9 (#95)
- Qwen2.5 7B Instruct: 36.5 (#208)

| Benchmark | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| DeepSWE | 30.5% | — |
| FrontierCode | 30.1% | — |
| LMArena WebDev | 1473 | — |
| SciCode | 47.5% | — |
| WeirdML | 54.1% | — |
| BigCodeBench Instruct | — | 37.6% |
| BigCodeBench Complete | — | 46.1% |
| ALE-Bench | 886.23 | — |

## Agentic & Tool Use

- Kimi K2.7 Code: 24.0 (#122)
- Qwen2.5 7B Instruct: 23.8 (#124)

| Benchmark | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| APEX-Agents | 37.6% | — |
| BALROG | — | 7.8% |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 5,083 | — |

## Reasoning

- Kimi K2.7 Code: 39.0 (#61)
- Qwen2.5 7B Instruct: 14.8 (#322)

| Benchmark | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| Chess Puzzles | 21% | 0% |
| Epoch Capabilities Index | 149.97 | 118.51 |
| SimpleBench | 57.9% | — |
| CritPt | 10% | — |
| DTBench | — | 47.7% |
| LMCA | — | 6.4% |
| Surface Evolver Bench | 48.8% | — |

## Math

- Kimi K2.7 Code: 52.9 (#48)
- Qwen2.5 7B Instruct: 12.6 (#306)

| Benchmark | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 95.6% | 2.5% |
| FrontierMath (Tiers 1-3) | 54% | — |
| FrontierMath Tier 4 | 12.2% | — |
| Omni-MATH | — | 29.4% |

## Knowledge

- Kimi K2.7 Code: 53.5 (#57)
- Qwen2.5 7B Instruct: 17.0 (#286)

| Benchmark | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 87.9% | 35.5% |
| SimpleQA Verified | 36.5% | — |
| MMLU-Pro | — | 53.9% |
| GPQA (HELM) | — | 34.1% |
| MMLU | — | 72.9% |

## Instruction Following

- Kimi K2.7 Code: —
- Qwen2.5 7B Instruct: 63.2 (#231)

| Benchmark | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | — | 74.1% |

## Writing & Preference

- Kimi K2.7 Code: —
- Qwen2.5 7B Instruct: 48.8 (#195)

| Benchmark | Kimi K2.7 Code | Qwen2.5 7B Instruct |
|---|---|---|
| WildBench | — | 73.1% |

## FAQ

### Is Kimi K2.7 Code better than Qwen2.5 7B Instruct?

Kimi K2.7 Code is the stronger model overall, scoring 43.3 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 5.6× less per token, which makes it the better buy when Kimi K2.7 Code's lead doesn't matter for your workload.

### Which is cheaper, Kimi K2.7 Code or Qwen2.5 7B Instruct?

Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Kimi K2.7 Code lists at $0.95 and $4.

### Is Kimi K2.7 Code or Qwen2.5 7B Instruct better for coding?

Kimi K2.7 Code scores higher on coding benchmarks: 42.9 versus 36.5 in the Noometry coding category.

### Which has the bigger context window?

Kimi K2.7 Code does, with 262K tokens against 131K.

### How many benchmarks do Kimi K2.7 Code and Qwen2.5 7B Instruct share?

4 benchmarks have published results for both models. Kimi K2.7 Code has 19 scored results on Noometry and Qwen2.5 7B Instruct has 15.