Model comparison
Command R vs DeepSeek-V2 (MoE-236B, May 2024)
Command R has enough public results to be ranked (#267); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 3 shared benchmarks.
Cohere
DeepSeekSummary
- They share 3 benchmarks with published results for both. Command R scores higher in 0 categories and DeepSeek-V2 (MoE-236B, May 2024) in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 39.8 to 28.8.
- The biggest single-benchmark swing is BigCodeBench Complete: 45.2% for Command R and 59.4% for DeepSeek-V2 (MoE-236B, May 2024).
Side by side
| Command R | DeepSeek-V2 (MoE-236B, May 2024) | |
|---|---|---|
| Provider | Cohere | DeepSeek |
| Noometry Index | 31.0 | 39.7 |
| Released | 2024-08-30 | 2024-05-07 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 29 | 10 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2 (MoE-236B, May 2024) leads
Command R: 28.8 (#301), DeepSeek-V2 (MoE-236B, May 2024): 39.8 (#141)
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| BigCodeBench Instruct | 37.1% | 48.9% |
| BigCodeBench Complete | 45.2% | 59.4% |
| LiveBench Coding | 17.9% | — |
| LMArena Coding | 1170 | — |
Reasoning Not comparable
Command R: 13.3 (#329), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LiveBench Reasoning | 21.9% | — |
| LMArena Hard Prompts | 1164 | — |
| DTBench | 46.4% | — |
| LiveBench Data Analysis | 33.3% | — |
| LMCA | 9.2% | — |
| BIG-Bench Hard | — | 78.8% |
| Epoch Capabilities Index | — | 124.78 |
| HellaSwag | — | 87.1% |
| LiveBench | 27.5% | — |
| PIQA | — | 83.9% |
| WinoGrande | — | 86.3% |
Math Not comparable
Command R: 25.8 (#244), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LiveBench Math | 19.4% | — |
| LMArena Math | 1155 | — |
Knowledge Not comparable
Command R: 30.5 (#219), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| MMLU | 65.2% | 78.4% |
| LMArena Expert | 1138 | — |
| ARC (AI2) Challenge | — | 92.2% |
| TriviaQA | — | 80% |
Multilingual Not comparable
Command R: 35.9 (#240), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LMArena Non-English | 1174 | — |
| LMArena Chinese | 1182 | — |
| LMArena French | 1161 | — |
| LMArena German | 1176 | — |
| LMArena Japanese | 1143 | — |
| LMArena Korean | 1162 | — |
| LMArena Russian | 1174 | — |
| LMArena Spanish | 1150 | — |
Instruction Following Not comparable
Command R: 58.3 (#256), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LiveBench Instruction Following | 55.6% | — |
| LMArena Instruction Following | 1167 | — |
Long Context Not comparable
Command R: 36.3 (#226), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LMArena Longer Query | 1198 | — |
Writing & Preference Not comparable
Command R: 38.4 (#248), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Command R | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LMArena Text | 1187 | — |
| LMArena Creative Writing | 1170 | — |
| LMArena Multi-Turn | 1163 | — |
| LiveBench Language | 16.7% | — |
Frequently asked questions
Is Command R better than DeepSeek-V2 (MoE-236B, May 2024)?
Command R has enough public results to be ranked (#267); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is Command R or DeepSeek-V2 (MoE-236B, May 2024) better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 39.8 versus 28.8 in the Noometry coding category.
How many benchmarks do Command R and DeepSeek-V2 (MoE-236B, May 2024) share?
3 benchmarks have published results for both models. Command R has 29 scored results on Noometry and DeepSeek-V2 (MoE-236B, May 2024) has 10.