Model comparison
Llama 2-13B vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 29.6 on the Noometry Index.
Last verified . 15 shared benchmarks.
Meta
StepFunSummary
- They share 15 benchmarks with published results for both. Llama 2-13B 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 29.8.
Side by side
| Llama 2-13B | Step 3 | |
|---|---|---|
| Provider | Meta | StepFun |
| Noometry Index | 29.6 | 40.5 |
| Released | 2023-07-18 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 32 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Llama 2-13B: 30.9 (#291), Step 3: 40.1 (#147)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Coding | 1062 | 1367 |
Reasoning Step 3 leads
Llama 2-13B: 12.8 (#337), Step 3: 28.4 (#105)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1051 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| Chess Puzzles | 0% | — |
| DTBench | 42.2% | — |
| BIG-Bench Hard | 58.2% | — |
| Epoch Capabilities Index | 106.17 | — |
| HellaSwag | 80.7% | — |
| LAMBADA | 76.5% | — |
| PIQA | 80.8% | — |
| WinoGrande | 72.8% | — |
Math Step 3 leads
Llama 2-13B: 31.1 (#229), Step 3: 37.6 (#148)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Math | 1065 | 1366 |
| GSM8K | 36.9% | — |
Knowledge Step 3 leads
Llama 2-13B: 28.1 (#249), Step 3: 36.8 (#164)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Expert | 1030 | 1333 |
| ARC (AI2) Challenge | 60.3% | — |
| BoolQ | 82.4% | — |
| MMLU | 55.6% | — |
| OpenBookQA | 57% | — |
| TriviaQA | 79.6% | — |
Multimodal Not comparable
Llama 2-13B: —, Step 3: 35.5 (#86)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
| ScienceQA | 55.8% | — |
Multilingual Step 3 leads
Llama 2-13B: 26.5 (#279), Step 3: 46.3 (#159)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Non-English | 1024 | 1327 |
| LMArena Chinese | 1001 | 1397 |
| LMArena German | 1009 | 1371 |
| LMArena Korean | 953 | 1269 |
| LMArena Russian | 1055 | 1331 |
| LMArena Spanish | 1087 | 1371 |
| LMArena French | 1044 | — |
| LMArena Japanese | 894 | — |
Instruction Following Step 3 leads
Llama 2-13B: 53.3 (#287), Step 3: 70.4 (#164)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1045 | 1332 |
Long Context Step 3 leads
Llama 2-13B: 32.3 (#269), Step 3: 40.3 (#157)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Longer Query | 1064 | 1326 |
Writing & Preference Step 3 leads
Llama 2-13B: 29.8 (#289), Step 3: 54.3 (#151)
| Benchmark | Llama 2-13B | Step 3 |
|---|---|---|
| LMArena Text | 1084 | 1350 |
| LMArena Creative Writing | 1047 | 1321 |
| LMArena Multi-Turn | 1050 | 1341 |
Frequently asked questions
Is Llama 2-13B better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 29.6 on the Noometry Index.
Is Llama 2-13B or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 30.9 in the Noometry coding category.
How many benchmarks do Llama 2-13B and Step 3 share?
15 benchmarks have published results for both models. Llama 2-13B has 32 scored results on Noometry and Step 3 has 17.