Model comparison
Command R+ vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 32.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
Cohere
StepFunSummary
- They share 15 benchmarks with published results for both. Command R+ scores higher in 0 categories and Step 3 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3 leads 28.4 to 9.2.
Side by side
| Command R+ | Step 3 | |
|---|---|---|
| Provider | Cohere | StepFun |
| Noometry Index | 32.4 | 40.5 |
| Released | 2024-08-30 | — |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 34 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Command R+: 29.1 (#309), Step 3: 40.1 (#147)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Coding | 1187 | 1367 |
| BigCodeBench Instruct | 33.8% | — |
| LiveBench Coding | 19.1% | — |
| BigCodeBench Complete | 41.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 63.5% | — |
Reasoning Step 3 leads
Command R+: 9.2 (#344), Step 3: 28.4 (#105)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1186 | 1355 |
| SimpleBench | 17.4% | — |
| Kagi LLM Benchmark | — | 62.3% |
| LiveBench Reasoning | 24.8% | — |
| DTBench | 54.9% | — |
| LiveBench Data Analysis | 38.1% | — |
| LMCA | 5% | — |
| Epoch Capabilities Index | 119.34 | — |
| LiveBench | 31.8% | — |
Math Step 3 leads
Command R+: 28.9 (#242), Step 3: 37.6 (#148)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Math | 1188 | 1366 |
| LiveBench Math | 21.3% | — |
Knowledge Too close to call
Command R+: 36.4 (#169), Step 3: 36.8 (#164)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Expert | 1174 | 1333 |
| Vectara Hallucination Rate | 6.9% | — |
| MMLU | 69.4% | — |
Multimodal Not comparable
Command R+: —, Step 3: 35.5 (#86)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Step 3 leads
Command R+: 38.6 (#227), Step 3: 46.3 (#159)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Non-English | 1216 | 1327 |
| LMArena Chinese | 1226 | 1397 |
| LMArena German | 1216 | 1371 |
| LMArena Korean | 1138 | 1269 |
| LMArena Russian | 1227 | 1331 |
| LMArena Spanish | 1189 | 1371 |
| LMArena French | 1209 | — |
| LMArena Japanese | 1166 | — |
Instruction Following Step 3 leads
Command R+: 60.0 (#254), Step 3: 70.4 (#164)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1197 | 1332 |
| LiveBench Instruction Following | 57.6% | — |
Long Context Step 3 leads
Command R+: 37.3 (#219), Step 3: 40.3 (#157)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Longer Query | 1230 | 1326 |
Writing & Preference Step 3 leads
Command R+: 43.5 (#228), Step 3: 54.3 (#151)
| Benchmark | Command R+ | Step 3 |
|---|---|---|
| LMArena Text | 1229 | 1350 |
| LMArena Creative Writing | 1235 | 1321 |
| LMArena Multi-Turn | 1213 | 1341 |
| LiveBench Language | 29.7% | — |
Frequently asked questions
Is Command R+ better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 32.4 on the Noometry Index.
Is Command R+ or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 29.1 in the Noometry coding category.
How many benchmarks do Command R+ and Step 3 share?
15 benchmarks have published results for both models. Command R+ has 34 scored results on Noometry and Step 3 has 17.