# Deepseek Coder v2 vs Step 3.5 Flash

> Step 3.5 Flash is the stronger model overall, scoring 42.3 to 35.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-coder-v2-vs-step-3-5-flash
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and Step 3.5 Flash in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 3.5 Flash leads 58.8 to 38.2.

## Snapshot

| | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 35.9 | 42.3 |
| Rank | 220 | 116 |
| Context | — | 256K |
| Input $/M | — | $0.10 |
| Output $/M | — | $0.30 |
| Weights | Open | Open |

## Coding

- Deepseek Coder v2: 38.1 (#183)
- Step 3.5 Flash: 42.4 (#105)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1251 | 1436 |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |

## Reasoning

- Deepseek Coder v2: 23.6 (#176)
- Step 3.5 Flash: 22.2 (#202)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1411 |
| NYT Connections (extended) | — | 28.4% |
| WinoGrande | 83.7% | — |

## Math

- Deepseek Coder v2: 34.9 (#190)
- Step 3.5 Flash: 42.6 (#84)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1241 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
| GSM8K | 94.5% | — |

## Knowledge

- Deepseek Coder v2: 32.3 (#212)
- Step 3.5 Flash: 39.6 (#132)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1181 | 1421 |
| ARC (AI2) Challenge | 64.3% | — |

## Multilingual

- Deepseek Coder v2: 36.3 (#240)
- Step 3.5 Flash: 50.5 (#119)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1182 | 1385 |
| LMArena Chinese | 1201 | 1447 |
| LMArena French | 1185 | 1421 |
| LMArena German | 1164 | 1405 |
| LMArena Japanese | 1126 | 1354 |
| LMArena Korean | 1104 | 1352 |
| LMArena Russian | 1188 | 1385 |
| LMArena Spanish | 1153 | 1419 |

## Instruction Following

- Deepseek Coder v2: 61.7 (#242)
- Step 3.5 Flash: 73.1 (#124)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1180 | 1385 |

## Long Context

- Deepseek Coder v2: 37.0 (#224)
- Step 3.5 Flash: 42.8 (#117)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1219 | 1402 |

## Writing & Preference

- Deepseek Coder v2: 38.2 (#253)
- Step 3.5 Flash: 58.8 (#113)

| Benchmark | Deepseek Coder v2 | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1191 | 1403 |
| LMArena Creative Writing | 1120 | 1357 |
| LMArena Multi-Turn | 1177 | 1405 |

## FAQ

### Is Deepseek Coder v2 better than Step 3.5 Flash?

Step 3.5 Flash is the stronger model overall, scoring 42.3 to 35.9 on the Noometry Index.

### Is Deepseek Coder v2 or Step 3.5 Flash better for coding?

Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 38.1 in the Noometry coding category.

### How many benchmarks do Deepseek Coder v2 and Step 3.5 Flash share?

17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Step 3.5 Flash has 19.