Model comparison
Command R+ vs Llama 2-13B
Command R+ is the stronger model overall, scoring 32.4 to 29.6 on the Noometry Index.
Last verified . 20 shared benchmarks.
Cohere
MetaSummary
- They share 20 benchmarks with published results for both. Command R+ scores higher in 5 categories and Llama 2-13B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Command R+ leads 43.5 to 29.8.
- The biggest single-benchmark swing is DTBench: 54.9% for Command R+ and 42.2% for Llama 2-13B.
Side by side
| Command R+ | Llama 2-13B | |
|---|---|---|
| Provider | Cohere | Meta |
| Noometry Index | 32.4 | 29.6 |
| Released | 2024-08-30 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 34 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-13B leads
Command R+: 29.1 (#309), Llama 2-13B: 30.9 (#291)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1187 | 1062 |
| BigCodeBench Instruct | 33.8% | — |
| LiveBench Coding | 19.1% | — |
| BigCodeBench Complete | 41.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 63.5% | — |
Reasoning Llama 2-13B leads
Command R+: 9.2 (#344), Llama 2-13B: 12.8 (#337)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Hard Prompts | 1186 | 1051 |
| DTBench | 54.9% | 42.2% |
| Epoch Capabilities Index | 119.34 | 106.17 |
| SimpleBench | 17.4% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | 24.8% | — |
| LiveBench Data Analysis | 38.1% | — |
| LMCA | 5% | — |
| BIG-Bench Hard | — | 58.2% |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| LiveBench | 31.8% | — |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |
Math Llama 2-13B leads
Command R+: 28.9 (#242), Llama 2-13B: 31.1 (#229)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Math | 1188 | 1065 |
| LiveBench Math | 21.3% | — |
| GSM8K | — | 36.9% |
Knowledge Command R+ leads
Command R+: 36.4 (#169), Llama 2-13B: 28.1 (#249)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1174 | 1030 |
| MMLU | 69.4% | 55.6% |
| Vectara Hallucination Rate | 6.9% | — |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
Command R+: —, Llama 2-13B: —
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| ScienceQA | — | 55.8% |
Multilingual Command R+ leads
Command R+: 38.6 (#227), Llama 2-13B: 26.5 (#279)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1216 | 1024 |
| LMArena Chinese | 1226 | 1001 |
| LMArena French | 1209 | 1044 |
| LMArena German | 1216 | 1009 |
| LMArena Japanese | 1166 | 894 |
| LMArena Korean | 1138 | 953 |
| LMArena Russian | 1227 | 1055 |
| LMArena Spanish | 1189 | 1087 |
Instruction Following Command R+ leads
Command R+: 60.0 (#254), Llama 2-13B: 53.3 (#287)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1197 | 1045 |
| LiveBench Instruction Following | 57.6% | — |
Long Context Command R+ leads
Command R+: 37.3 (#219), Llama 2-13B: 32.3 (#269)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1230 | 1064 |
Writing & Preference Command R+ leads
Command R+: 43.5 (#228), Llama 2-13B: 29.8 (#289)
| Benchmark | Command R+ | Llama 2-13B |
|---|---|---|
| LMArena Text | 1229 | 1084 |
| LMArena Creative Writing | 1235 | 1047 |
| LMArena Multi-Turn | 1213 | 1050 |
| LiveBench Language | 29.7% | — |
Frequently asked questions
Is Command R+ better than Llama 2-13B?
Command R+ is the stronger model overall, scoring 32.4 to 29.6 on the Noometry Index.
Is Command R+ or Llama 2-13B better for coding?
Llama 2-13B scores higher on coding benchmarks: 30.9 versus 29.1 in the Noometry coding category.
How many benchmarks do Command R+ and Llama 2-13B share?
20 benchmarks have published results for both models. Command R+ has 34 scored results on Noometry and Llama 2-13B has 32.