Model comparison
Command R+ vs Llama 3-70B
Command R+ is the stronger model overall, scoring 32.0 to 28.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
Cohere
MetaSummary
- They share 24 benchmarks with published results for both. Command R+ scores higher in 5 categories and Llama 3-70B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Command R+ leads 35.9 to 20.8.
- The biggest single-benchmark swing is BigCodeBench Complete: 41.9% for Command R+ and 54.5% for Llama 3-70B.
Side by side
| Command R+ | Llama 3-70B | |
|---|---|---|
| Provider | Cohere | Meta |
| Noometry Index | 32.0 | 28.5 |
| Released | 2024-08-30 | 2024-04-18 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 34 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-70B leads
Command R+: 28.6 (#304), Llama 3-70B: 35.3 (#215)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| BigCodeBench Instruct | 33.8% | 43.6% |
| LMArena Coding | 1187 | 1206 |
| BigCodeBench Complete | 41.9% | 54.5% |
| HumanEval+ | 56.7% | 72% |
| MBPP+ | 63.5% | 69% |
| LiveBench Coding | 19.1% | — |
Agentic & Tool Use Not comparable
Command R+: —, Llama 3-70B: 20.9 (#132)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| Cybench | — | 5% |
Reasoning Llama 3-70B leads
Command R+: 8.9 (#341), Llama 3-70B: 17.4 (#284)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1186 | 1195 |
| DTBench | 54.9% | 54.2% |
| Epoch Capabilities Index | 119.28 | 122.91 |
| SimpleBench | 17.4% | — |
| Kagi LLM Benchmark | — | 35.1% |
| LiveBench Reasoning | 24.8% | — |
| LiveBench Data Analysis | 38.1% | — |
| LMCA | 5% | — |
| ForecastBench | — | 57.1 |
| LiveBench | 31.8% | — |
| WinoGrande | — | 83.5% |
Math Command R+ leads
Command R+: 26.6 (#239), Llama 3-70B: 12.0 (#299)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| LMArena Math | 1188 | 1218 |
| OTIS Mock AIME 2024-2025 | — | 4.3% |
| LiveBench Math | 21.3% | — |
| MATH Level 5 | — | 22.6% |
Knowledge Command R+ leads
Command R+: 35.9 (#165), Llama 3-70B: 20.8 (#272)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| LMArena Expert | 1174 | 1149 |
| MMLU | 69.4% | 79.3% |
| GPQA Diamond | — | 40.6% |
| Vectara Hallucination Rate | 6.9% | — |
Multilingual Command R+ leads
Command R+: 38.8 (#222), Llama 3-70B: 33.8 (#246)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1216 | 1142 |
| LMArena Chinese | 1226 | 1114 |
| LMArena French | 1208 | 1231 |
| LMArena German | 1216 | 1169 |
| LMArena Japanese | 1166 | 1017 |
| LMArena Korean | 1138 | 1017 |
| LMArena Russian | 1227 | 1159 |
| LMArena Spanish | 1189 | 1241 |
Instruction Following Llama 3-70B leads
Command R+: 60.2 (#249), Llama 3-70B: 62.8 (#233)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1197 | 1194 |
| LiveBench Instruction Following | 57.6% | — |
Long Context Command R+ leads
Command R+: 37.3 (#214), Llama 3-70B: 35.6 (#235)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1231 | 1174 |
Writing & Preference Too close to call
Command R+: 43.7 (#223), Llama 3-70B: 43.0 (#225)
| Benchmark | Command R+ | Llama 3-70B |
|---|---|---|
| LMArena Text | 1229 | 1221 |
| LMArena Creative Writing | 1235 | 1210 |
| LMArena Multi-Turn | 1213 | 1223 |
| LiveBench Language | 29.7% | — |
Frequently asked questions
Is Command R+ better than Llama 3-70B?
Command R+ is the stronger model overall, scoring 32.0 to 28.5 on the Noometry Index.
Is Command R+ or Llama 3-70B better for coding?
Llama 3-70B scores higher on coding benchmarks: 35.3 versus 28.6 in the Noometry coding category.
How many benchmarks do Command R+ and Llama 3-70B share?
24 benchmarks have published results for both models. Command R+ has 34 scored results on Noometry and Llama 3-70B has 31.