# Deepseek Coder v2 vs Phi 3 Mini 128k Instruct

> Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 29.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-coder-v2-vs-phi-3-mini-128k-instruct
- Last updated: 2026-10-11
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Deepseek Coder v2 scores higher in 8 categories and Phi 3 Mini 128k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Deepseek Coder v2 leads 38.2 to 27.1.
- The biggest single-benchmark swing is BigCodeBench Complete: 59.7% for Deepseek Coder v2 and 40.6% for Phi 3 Mini 128k Instruct.

## Snapshot

| | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 35.9 | 29.7 |
| Rank | 220 | 305 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Deepseek Coder v2: 38.1 (#183)
- Phi 3 Mini 128k Instruct: 28.8 (#312)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| BigCodeBench Instruct | 48.2% | 29.6% |
| LMArena Coding | 1251 | 1039 |
| BigCodeBench Complete | 59.7% | 40.6% |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |

## Reasoning

- Deepseek Coder v2: 23.6 (#176)
- Phi 3 Mini 128k Instruct: 19.6 (#256)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1028 |
| WinoGrande | 83.7% | — |

## Math

- Deepseek Coder v2: 34.9 (#190)
- Phi 3 Mini 128k Instruct: 31.6 (#222)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| LMArena Math | 1241 | 1089 |
| GSM8K | 94.5% | — |

## Knowledge

- Deepseek Coder v2: 32.3 (#212)
- Phi 3 Mini 128k Instruct: 26.8 (#254)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| LMArena Expert | 1181 | 984 |
| ARC (AI2) Challenge | 64.3% | — |

## Multilingual

- Deepseek Coder v2: 36.3 (#240)
- Phi 3 Mini 128k Instruct: 25.2 (#285)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| LMArena Non-English | 1182 | 1000 |
| LMArena Chinese | 1201 | 1016 |
| LMArena French | 1185 | 1039 |
| LMArena German | 1164 | 1006 |
| LMArena Japanese | 1126 | 899 |
| LMArena Korean | 1104 | 856 |
| LMArena Russian | 1188 | 1004 |
| LMArena Spanish | 1153 | 1059 |

## Instruction Following

- Deepseek Coder v2: 61.7 (#242)
- Phi 3 Mini 128k Instruct: 51.8 (#294)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| LMArena Instruction Following | 1180 | 1023 |

## Long Context

- Deepseek Coder v2: 37.0 (#224)
- Phi 3 Mini 128k Instruct: 30.4 (#289)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| LMArena Longer Query | 1219 | 996 |

## Writing & Preference

- Deepseek Coder v2: 38.2 (#253)
- Phi 3 Mini 128k Instruct: 27.1 (#301)

| Benchmark | Deepseek Coder v2 | Phi 3 Mini 128k Instruct |
|---|---|---|
| LMArena Text | 1191 | 1050 |
| LMArena Creative Writing | 1120 | 1024 |
| LMArena Multi-Turn | 1177 | 989 |

## FAQ

### Is Deepseek Coder v2 better than Phi 3 Mini 128k Instruct?

Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 29.7 on the Noometry Index.

### Is Deepseek Coder v2 or Phi 3 Mini 128k Instruct better for coding?

Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 28.8 in the Noometry coding category.

### How many benchmarks do Deepseek Coder v2 and Phi 3 Mini 128k Instruct share?

19 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Phi 3 Mini 128k Instruct has 19.