Model comparison
Command R+ vs DeepSeek-V3
DeepSeek-V3 is the stronger model overall, scoring 39.5 to 32.4 on the Noometry Index.
Last verified . 34 shared benchmarks.
Cohere
DeepSeekSummary
- They share 34 benchmarks with published results for both. Command R+ scores higher in 1 category and DeepSeek-V3 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3 leads 57.4 to 43.5.
- The biggest single-benchmark swing is LiveBench Math: 21.3% for Command R+ and 73.5% for DeepSeek-V3.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $2.50 / $10 for Command R+.
- DeepSeek-V3 accepts more context: 164K tokens versus 128K.
Side by side
| Command R+ | DeepSeek-V3 | |
|---|---|---|
| Provider | Cohere | DeepSeek |
| Noometry Index | 32.4 | 39.5 |
| Released | 2024-08-30 | 2024-12-26 |
| Weights | Open | Open |
| Context window | 128K | 164K |
| Max output | 4K | 164K |
| Input $ / M tokens | $2.50 | $0.24 |
| Output $ / M tokens | $10 | $0.90 |
| Results tracked | 34 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3 leads
Command R+: 29.1 (#309), DeepSeek-V3: 42.3 (#106)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| BigCodeBench Instruct | 33.8% | 50% |
| LiveBench Coding | 19.1% | 70.9% |
| LMArena Coding | 1187 | 1368 |
| BigCodeBench Complete | 41.9% | 62.2% |
| HumanEval+ | 56.7% | 86.6% |
| MBPP+ | 63.5% | 73% |
| Aider Polyglot | — | 55.1% |
| SciCode | — | 35.8% |
| WeirdML | — | 36.1% |
Agentic & Tool Use Not comparable
Command R+: —, DeepSeek-V3: —
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| METR Time Horizons | — | 49.6% |
Reasoning DeepSeek-V3 leads
Command R+: 9.2 (#344), DeepSeek-V3: 20.5 (#236)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| SimpleBench | 17.4% | 27.2% |
| LiveBench Reasoning | 24.8% | 65.8% |
| LMArena Hard Prompts | 1186 | 1365 |
| DTBench | 54.9% | 64.8% |
| LiveBench Data Analysis | 38.1% | 60.9% |
| LMCA | 5% | 15.5% |
| Epoch Capabilities Index | 119.34 | 135.94 |
| LiveBench | 31.8% | 66.9% |
| Kagi LLM Benchmark | — | 52.3% |
| CritPt | — | 0% |
| BIG-Bench Hard | — | 87.5% |
| ForecastBench | — | 59.1 |
| HellaSwag | — | 88.9% |
| PIQA | — | 84.7% |
| WinoGrande | — | 85.2% |
Math DeepSeek-V3 leads
Command R+: 28.9 (#242), DeepSeek-V3: 32.1 (#219)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| LiveBench Math | 21.3% | 73.5% |
| LMArena Math | 1188 | 1373 |
| OTIS Mock AIME 2024-2025 | — | 37.8% |
| Omni-MATH | — | 40.3% |
| MATH Level 5 | — | 75.5% |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge DeepSeek-V3 leads
Command R+: 36.4 (#169), DeepSeek-V3: 37.5 (#155)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| Vectara Hallucination Rate | 6.9% | 6.1% |
| LMArena Expert | 1174 | 1351 |
| MMLU | 69.4% | 87.2% |
| GPQA Diamond | — | 67.6% |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 26.1% |
| GPQA (HELM) | — | 53.8% |
| ARC (AI2) Challenge | — | 95.3% |
| TriviaQA | — | 82.9% |
Multilingual DeepSeek-V3 leads
Command R+: 38.6 (#227), DeepSeek-V3: 48.5 (#143)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| LMArena Non-English | 1216 | 1358 |
| LMArena Chinese | 1226 | 1391 |
| LMArena French | 1209 | 1385 |
| LMArena German | 1216 | 1374 |
| LMArena Japanese | 1166 | 1333 |
| LMArena Korean | 1138 | 1319 |
| LMArena Russian | 1227 | 1373 |
| LMArena Spanish | 1189 | 1358 |
Instruction Following DeepSeek-V3 leads
Command R+: 60.0 (#254), DeepSeek-V3: 72.8 (#130)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| LiveBench Instruction Following | 57.6% | 81.5% |
| LMArena Instruction Following | 1197 | 1345 |
| IFEval | — | 83.2% |
Long Context Command R+ leads
Command R+: 37.3 (#219), DeepSeek-V3: 34.0 (#253)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| LMArena Longer Query | 1230 | 1352 |
| Fiction.LiveBench | — | 50% |
Writing & Preference DeepSeek-V3 leads
Command R+: 43.5 (#228), DeepSeek-V3: 57.4 (#130)
| Benchmark | Command R+ | DeepSeek-V3 |
|---|---|---|
| LMArena Text | 1229 | 1375 |
| LMArena Creative Writing | 1235 | 1364 |
| LMArena Multi-Turn | 1213 | 1389 |
| LiveBench Language | 29.7% | 49.1% |
| Short-Story Creative Writing | — | 77% |
| EQ-Bench Creative Writing | — | 1472 |
| WildBench | — | 83% |
Frequently asked questions
Is Command R+ better than DeepSeek-V3?
DeepSeek-V3 is the stronger model overall, scoring 39.5 to 32.4 on the Noometry Index.
Which is cheaper, Command R+ or DeepSeek-V3?
DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Command R+ lists at $2.50 and $10.
Is Command R+ or DeepSeek-V3 better for coding?
DeepSeek-V3 scores higher on coding benchmarks: 42.3 versus 29.1 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3 does, with 164K tokens against 128K.
How many benchmarks do Command R+ and DeepSeek-V3 share?
34 benchmarks have published results for both models. Command R+ has 34 scored results on Noometry and DeepSeek-V3 has 60.