Model comparison
Command R+ vs Qwen2-72B
Command R+ is the stronger model overall, scoring 32.4 to 30.0 on the Noometry Index.
Last verified . 21 shared benchmarks.
Cohere
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. Command R+ scores higher in 5 categories and Qwen2-72B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Command R+ leads 36.4 to 21.2.
- The biggest single-benchmark swing is BigCodeBench Complete: 41.9% for Command R+ and 54% for Qwen2-72B.
Side by side
| Command R+ | Qwen2-72B | |
|---|---|---|
| Provider | Cohere | Alibaba (Qwen) |
| Noometry Index | 32.4 | 30.0 |
| Released | 2024-08-30 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 34 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Command R+: 29.1 (#309), Qwen2-72B: 29.1 (#310)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| BigCodeBench Instruct | 33.8% | 38.5% |
| LMArena Coding | 1187 | 1196 |
| BigCodeBench Complete | 41.9% | 54% |
| WeirdML | — | 11.3% |
| LiveBench Coding | 19.1% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 63.5% | — |
Agentic & Tool Use Not comparable
Command R+: —, Qwen2-72B: 17.0 (#146)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Command R+: 9.2 (#344), Qwen2-72B: 23.2 (#181)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1186 | 1191 |
| Epoch Capabilities Index | 119.34 | 125.28 |
| SimpleBench | 17.4% | — |
| LiveBench Reasoning | 24.8% | — |
| DTBench | 54.9% | — |
| LiveBench Data Analysis | 38.1% | — |
| LMCA | 5% | — |
| LiveBench | 31.8% | — |
Math Qwen2-72B leads
Command R+: 28.9 (#242), Qwen2-72B: 30.2 (#236)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| LMArena Math | 1188 | 1235 |
| LiveBench Math | 21.3% | — |
| MATH Level 5 | — | 39.1% |
Knowledge Command R+ leads
Command R+: 36.4 (#169), Qwen2-72B: 21.2 (#275)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| LMArena Expert | 1174 | 1171 |
| MMLU | 69.4% | 82.4% |
| GPQA Diamond | — | 40.8% |
| Vectara Hallucination Rate | 6.9% | — |
Multilingual Command R+ leads
Command R+: 38.6 (#227), Qwen2-72B: 35.9 (#244)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1216 | 1176 |
| LMArena Chinese | 1226 | 1240 |
| LMArena French | 1209 | 1170 |
| LMArena German | 1216 | 1151 |
| LMArena Japanese | 1166 | 1111 |
| LMArena Korean | 1138 | 1083 |
| LMArena Russian | 1227 | 1169 |
| LMArena Spanish | 1189 | 1169 |
Instruction Following Qwen2-72B leads
Command R+: 60.0 (#254), Qwen2-72B: 61.7 (#241)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1197 | 1181 |
| LiveBench Instruction Following | 57.6% | — |
Long Context Command R+ leads
Command R+: 37.3 (#219), Qwen2-72B: 36.1 (#235)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1230 | 1192 |
Writing & Preference Command R+ leads
Command R+: 43.5 (#228), Qwen2-72B: 40.8 (#241)
| Benchmark | Command R+ | Qwen2-72B |
|---|---|---|
| LMArena Text | 1229 | 1203 |
| LMArena Creative Writing | 1235 | 1181 |
| LMArena Multi-Turn | 1213 | 1196 |
| LiveBench Language | 29.7% | — |
Frequently asked questions
Is Command R+ better than Qwen2-72B?
Command R+ is the stronger model overall, scoring 32.4 to 30.0 on the Noometry Index.
Is Command R+ or Qwen2-72B better for coding?
They score almost the same on coding (29.1 vs 29.1); test both on your own repository before choosing.
How many benchmarks do Command R+ and Qwen2-72B share?
21 benchmarks have published results for both models. Command R+ has 34 scored results on Noometry and Qwen2-72B has 26.