# Kimi K2.7 Code vs Phi-4

> Kimi K2.7 Code is the stronger model overall, scoring 43.0 to 30.9 on the Noometry Index. Phi-4 costs 20× 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-phi-4
- Last updated: 2026-10-10
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. Kimi K2.7 Code scores higher in 5 categories and Phi-4 in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.7 Code leads 52.6 to 19.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 95.6% for Kimi K2.7 Code and 13.8% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.95 / $4 for Kimi K2.7 Code.
- Kimi K2.7 Code accepts more context: 262K tokens versus 128K.

## Snapshot

| | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| Provider | Moonshot AI | Microsoft |
| Noometry Index | 43.0 | 30.9 |
| Rank | 90 | 269 |
| Context | 262K | 128K |
| Input $/M | $0.95 | $0.07 |
| Output $/M | $4 | $0.14 |
| Weights | Open | Open |

## Coding

- Kimi K2.7 Code: 42.4 (#92)
- Phi-4: 33.8 (#237)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| DeepSWE | 30.5% | — |
| FrontierCode | 30.1% | — |
| LMArena WebDev | 1473 | — |
| SciCode | 47.5% | — |
| WeirdML | 54.1% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| LMArena Coding | — | 1231 |
| BigCodeBench Complete | — | 55.4% |
| ALE-Bench | 886.23 | — |

## Agentic & Tool Use

- Kimi K2.7 Code: 23.9 (#115)
- Phi-4: 22.6 (#120)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| APEX-Agents | 37.6% | — |
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 5,083 | — |

## Reasoning

- Kimi K2.7 Code: 38.2 (#56)
- Phi-4: 17.0 (#287)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| Chess Puzzles | 21% | 1% |
| Epoch Capabilities Index | 150.01 | 130.43 |
| SimpleBench | 57.9% | — |
| CritPt | 10% | — |
| LiveBench Reasoning | — | 47.8% |
| LMArena Hard Prompts | — | 1220 |
| LiveBench Data Analysis | — | 45.2% |
| Surface Evolver Bench | 48.8% | — |
| LiveBench | — | 41.6% |

## Math

- Kimi K2.7 Code: 52.6 (#39)
- Phi-4: 19.6 (#279)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 95.6% | 13.8% |
| FrontierMath (Tiers 1-3) | 54% | — |
| FrontierMath Tier 4 | 12.2% | — |
| LiveBench Math | — | 42% |
| LMArena Math | — | 1246 |
| MATH Level 5 | — | 64.9% |

## Knowledge

- Kimi K2.7 Code: 53.6 (#51)
- Phi-4: 32.5 (#204)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| GPQA Diamond | 87.9% | 56.1% |
| SimpleQA Verified | 36.5% | — |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| LMArena Expert | — | 1203 |
| MMLU | — | 84.8% |

## Multilingual

- Kimi K2.7 Code: —
- Phi-4: 37.4 (#232)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| LMArena Non-English | — | 1197 |
| LMArena Chinese | — | 1211 |
| LMArena French | — | 1223 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1158 |
| LMArena Korean | — | 1151 |
| LMArena Russian | — | 1209 |
| LMArena Spanish | — | 1234 |

## Instruction Following

- Kimi K2.7 Code: —
- Phi-4: 60.6 (#246)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| LiveBench Instruction Following | — | 58.4% |
| LMArena Instruction Following | — | 1201 |

## Long Context

- Kimi K2.7 Code: —
- Phi-4: 36.9 (#221)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| LMArena Longer Query | — | 1217 |

## Writing & Preference

- Kimi K2.7 Code: —
- Phi-4: 40.7 (#239)

| Benchmark | Kimi K2.7 Code | Phi-4 |
|---|---|---|
| LMArena Text | — | 1217 |
| LMArena Creative Writing | — | 1182 |
| Short-Story Creative Writing | — | 62.6% |
| LMArena Multi-Turn | — | 1206 |
| LiveBench Language | — | 25.6% |

## FAQ

### Is Kimi K2.7 Code better than Phi-4?

Kimi K2.7 Code is the stronger model overall, scoring 43.0 to 30.9 on the Noometry Index. Phi-4 costs 20× 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 Phi-4?

Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Kimi K2.7 Code lists at $0.95 and $4.

### Is Kimi K2.7 Code or Phi-4 better for coding?

Kimi K2.7 Code scores higher on coding benchmarks: 42.4 versus 33.8 in the Noometry coding category.

### Which has the bigger context window?

Kimi K2.7 Code does, with 262K tokens against 128K.

### How many benchmarks do Kimi K2.7 Code and Phi-4 share?

4 benchmarks have published results for both models. Kimi K2.7 Code has 19 scored results on Noometry and Phi-4 has 37.