# Command R+ vs DeepSeek-V3.2-Speciale

> DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 32.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/command-r-plus-vs-deepseek-v3-2-speciale
- Last updated: 2026-10-12
- Shared benchmarks: 1

## Summary

- They share 1 benchmark with published results for both. Command R+ scores higher in 0 categories and DeepSeek-V3.2-Speciale in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.2-Speciale leads 32.9 to 9.2.
- The biggest single-benchmark swing is SimpleBench: 17.4% for Command R+ and 52.6% for DeepSeek-V3.2-Speciale.
- DeepSeek-V3.2-Speciale is cheaper at $0.58 / $1.68 per million input/output tokens, against $2.50 / $10 for Command R+.

## Snapshot

| | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| Provider | Cohere | DeepSeek |
| Noometry Index | 32.4 | 39.7 |
| Rank | 257 | 162 |
| Context | 128K | 128K |
| Input $/M | $2.50 | $0.58 |
| Output $/M | $10 | $1.68 |
| Weights | Open | Open |

## Coding

- Command R+: 29.1 (#309)
- DeepSeek-V3.2-Speciale: 40.4 (#140)

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| WeirdML | — | 46.7% |
| BigCodeBench Instruct | 33.8% | — |
| LiveBench Coding | 19.1% | — |
| LMArena Coding | 1187 | — |
| BigCodeBench Complete | 41.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 63.5% | — |

## Reasoning

- Command R+: 9.2 (#344)
- DeepSeek-V3.2-Speciale: 32.9 (#73)

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| SimpleBench | 17.4% | 52.6% |
| LiveBench Reasoning | 24.8% | — |
| LMArena Hard Prompts | 1186 | — |
| DTBench | 54.9% | — |
| LiveBench Data Analysis | 38.1% | — |
| LMCA | 5% | — |
| Epoch Capabilities Index | 119.34 | — |
| LiveBench | 31.8% | — |

## Math

- Command R+: 28.9 (#242)
- DeepSeek-V3.2-Speciale: —

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| LiveBench Math | 21.3% | — |
| LMArena Math | 1188 | — |

## Knowledge

- Command R+: 36.4 (#169)
- DeepSeek-V3.2-Speciale: —

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| Vectara Hallucination Rate | 6.9% | — |
| LMArena Expert | 1174 | — |
| MMLU | 69.4% | — |

## Multilingual

- Command R+: 38.6 (#227)
- DeepSeek-V3.2-Speciale: —

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Non-English | 1216 | — |
| LMArena Chinese | 1226 | — |
| LMArena French | 1209 | — |
| LMArena German | 1216 | — |
| LMArena Japanese | 1166 | — |
| LMArena Korean | 1138 | — |
| LMArena Russian | 1227 | — |
| LMArena Spanish | 1189 | — |

## Instruction Following

- Command R+: 60.0 (#254)
- DeepSeek-V3.2-Speciale: —

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| LiveBench Instruction Following | 57.6% | — |
| LMArena Instruction Following | 1197 | — |

## Long Context

- Command R+: 37.3 (#219)
- DeepSeek-V3.2-Speciale: —

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Longer Query | 1230 | — |

## Writing & Preference

- Command R+: 43.5 (#228)
- DeepSeek-V3.2-Speciale: 46.0 (#222)

| Benchmark | Command R+ | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Text | 1229 | — |
| LMArena Creative Writing | 1235 | — |
| EQ-Bench Creative Writing | — | 1276 |
| LMArena Multi-Turn | 1213 | — |
| LiveBench Language | 29.7% | — |

## FAQ

### Is Command R+ better than DeepSeek-V3.2-Speciale?

DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 32.4 on the Noometry Index.

### Which is cheaper, Command R+ or DeepSeek-V3.2-Speciale?

DeepSeek-V3.2-Speciale is cheaper. It lists at $0.58 per million input tokens and $1.68 per million output tokens; Command R+ lists at $2.50 and $10.

### Is Command R+ or DeepSeek-V3.2-Speciale better for coding?

DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 40.4 versus 29.1 in the Noometry coding category.

### Which has the bigger context window?

Both accept 128K tokens.

### How many benchmarks do Command R+ and DeepSeek-V3.2-Speciale share?

1 benchmark has published results for both models. Command R+ has 34 scored results on Noometry and DeepSeek-V3.2-Speciale has 3.