# DeepSeek Coder 6.7B vs Phi-4

> Phi-4 has enough public results to be ranked (#269); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.

- Canonical page: https://noometry.com/compare/deepseek-coder-6-7b-vs-phi-4
- Last updated: 2026-10-10
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. DeepSeek Coder 6.7B scores higher in 1 category and Phi-4 in 0 categories; one gap is clear of the uncertainty.
- The biggest single-benchmark swing is BigCodeBench Complete: 43.8% for DeepSeek Coder 6.7B and 55.4% for Phi-4.

## Snapshot

| | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 37.1 | 30.9 |
| Rank | — | 269 |
| Context | — | 128K |
| Input $/M | — | $0.07 |
| Output $/M | — | $0.14 |
| Weights | Open | Open |

## Coding

- DeepSeek Coder 6.7B: 35.2 (#216)
- Phi-4: 33.8 (#237)

| Benchmark | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 35.5% | 45.5% |
| BigCodeBench Complete | 43.8% | 55.4% |
| LiveBench Coding | — | 30.7% |
| LMArena Coding | — | 1231 |
| HumanEval+ | 71.3% | — |
| MBPP+ | 65.6% | — |

## Agentic & Tool Use

- DeepSeek Coder 6.7B: —
- Phi-4: 22.6 (#120)

| Benchmark | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |

## Reasoning

- DeepSeek Coder 6.7B: —
- Phi-4: 17.0 (#287)

| Benchmark | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| Epoch Capabilities Index | 88.88 | 130.43 |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LMArena Hard Prompts | — | 1220 |
| LiveBench Data Analysis | — | 45.2% |
| LiveBench | — | 41.6% |
| WinoGrande | 57.6% | — |

## Math

- DeepSeek Coder 6.7B: —
- Phi-4: 19.6 (#279)

| Benchmark | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| LMArena Math | — | 1246 |
| MATH Level 5 | — | 64.9% |
| GSM8K | 21.3% | — |

## Knowledge

- DeepSeek Coder 6.7B: —
- Phi-4: 32.5 (#204)

| Benchmark | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| MMLU | 36.4% | 84.8% |
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| LMArena Expert | — | 1203 |
| ARC (AI2) Challenge | 36.4% | — |

## Multilingual

- DeepSeek Coder 6.7B: —
- Phi-4: 37.4 (#232)

| Benchmark | DeepSeek Coder 6.7B | 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

- DeepSeek Coder 6.7B: —
- Phi-4: 60.6 (#246)

| Benchmark | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| LiveBench Instruction Following | — | 58.4% |
| LMArena Instruction Following | — | 1201 |

## Long Context

- DeepSeek Coder 6.7B: —
- Phi-4: 36.9 (#221)

| Benchmark | DeepSeek Coder 6.7B | Phi-4 |
|---|---|---|
| LMArena Longer Query | — | 1217 |

## Writing & Preference

- DeepSeek Coder 6.7B: —
- Phi-4: 40.7 (#239)

| Benchmark | DeepSeek Coder 6.7B | 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 DeepSeek Coder 6.7B better than Phi-4?

Phi-4 has enough public results to be ranked (#269); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.

### Is DeepSeek Coder 6.7B or Phi-4 better for coding?

DeepSeek Coder 6.7B scores higher on coding benchmarks: 35.2 versus 33.8 in the Noometry coding category.

### How many benchmarks do DeepSeek Coder 6.7B and Phi-4 share?

4 benchmarks have published results for both models. DeepSeek Coder 6.7B has 9 scored results on Noometry and Phi-4 has 37.