Model comparison
Llama 2-7B vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 29.1 on the Noometry Index.
Last verified . 14 shared benchmarks.
Meta
StepFunSummary
- They share 14 benchmarks with published results for both. Llama 2-7B scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 3 leads 54.3 to 28.0.
Side by side
| Llama 2-7B | Step 3 | |
|---|---|---|
| Provider | Meta | StepFun |
| Noometry Index | 29.1 | 40.5 |
| Released | 2023-07-18 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 29 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Llama 2-7B: 29.2 (#307), Step 3: 40.1 (#147)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Coding | 1002 | 1367 |
Reasoning Step 3 leads
Llama 2-7B: 15.7 (#312), Step 3: 28.4 (#105)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1009 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| Chess Puzzles | 0% | — |
| BIG-Bench Hard | 39.2% | — |
| Epoch Capabilities Index | 99.06 | — |
| HellaSwag | 77.2% | — |
| LAMBADA | 73.3% | — |
| PIQA | 78.8% | — |
| WinoGrande | 69.2% | — |
Math Step 3 leads
Llama 2-7B: 30.7 (#233), Step 3: 37.6 (#148)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Math | 1042 | 1366 |
| GSM8K | 16.7% | — |
Knowledge Step 3 leads
Llama 2-7B: 28.2 (#248), Step 3: 36.8 (#164)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Expert | 1036 | 1333 |
| ARC (AI2) Challenge | 45.9% | — |
| BoolQ | 77.9% | — |
| MMLU | 45.8% | — |
| OpenBookQA | 58.6% | — |
| TriviaQA | 73.7% | — |
Multimodal Not comparable
Llama 2-7B: —, Step 3: 35.5 (#86)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
| ScienceQA | 43.1% | — |
Multilingual Step 3 leads
Llama 2-7B: 23.8 (#293), Step 3: 46.3 (#159)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Non-English | 973 | 1327 |
| LMArena Chinese | 973 | 1397 |
| LMArena German | 978 | 1371 |
| LMArena Russian | 995 | 1331 |
| LMArena Spanish | 1007 | 1371 |
| LMArena French | 970 | — |
| LMArena Korean | — | 1269 |
Instruction Following Step 3 leads
Llama 2-7B: 50.8 (#298), Step 3: 70.4 (#164)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1006 | 1332 |
Long Context Step 3 leads
Llama 2-7B: 30.4 (#287), Step 3: 40.3 (#157)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Longer Query | 999 | 1326 |
Writing & Preference Step 3 leads
Llama 2-7B: 28.0 (#298), Step 3: 54.3 (#151)
| Benchmark | Llama 2-7B | Step 3 |
|---|---|---|
| LMArena Text | 1053 | 1350 |
| LMArena Creative Writing | 1033 | 1321 |
| LMArena Multi-Turn | 1029 | 1341 |
Frequently asked questions
Is Llama 2-7B better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 29.1 on the Noometry Index.
Is Llama 2-7B or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 29.2 in the Noometry coding category.
How many benchmarks do Llama 2-7B and Step 3 share?
14 benchmarks have published results for both models. Llama 2-7B has 29 scored results on Noometry and Step 3 has 17.