Model comparison
Qwen1.5-32B vs Step 3
Step 3 is the stronger model overall, scoring 39.9 to 30.0 on the Noometry Index.
Last verified . 15 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 15 benchmarks with published results for both. Qwen1.5-32B scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Step 3 leads 36.2 to 13.5.
Side by side
| Qwen1.5-32B | Step 3 | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 30.0 | 39.9 |
| Released | 2024-02-04 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Qwen1.5-32B: 31.1 (#278), Step 3: 39.5 (#144)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Coding | 1155 | 1367 |
| BigCodeBench Instruct | 32.3% | — |
| BigCodeBench Complete | 42% | — |
Reasoning Step 3 leads
Qwen1.5-32B: 20.9 (#210), Step 3: 27.6 (#102)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1131 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
Math Step 3 leads
Qwen1.5-32B: 30.5 (#207), Step 3: 34.9 (#148)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Math | 1155 | 1366 |
Knowledge Step 3 leads
Qwen1.5-32B: 13.5 (#291), Step 3: 36.2 (#161)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Expert | 1126 | 1333 |
| GPQA Diamond | 30.7% | — |
| MMLU | 74.4% | — |
Multimodal Not comparable
Qwen1.5-32B: —, Step 3: 34.9 (#81)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |
Multilingual Step 3 leads
Qwen1.5-32B: 31.6 (#254), Step 3: 46.5 (#154)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Non-English | 1106 | 1327 |
| LMArena Chinese | 1176 | 1396 |
| LMArena German | 1058 | 1371 |
| LMArena Korean | 1007 | 1271 |
| LMArena Russian | 1073 | 1331 |
| LMArena Spanish | 1089 | 1372 |
| LMArena French | 1100 | — |
| LMArena Japanese | 1027 | — |
Instruction Following Step 3 leads
Qwen1.5-32B: 58.0 (#260), Step 3: 70.6 (#159)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1116 | 1332 |
Long Context Step 3 leads
Qwen1.5-32B: 34.7 (#241), Step 3: 40.3 (#152)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Longer Query | 1146 | 1326 |
Writing & Preference Step 3 leads
Qwen1.5-32B: 34.4 (#266), Step 3: 54.5 (#146)
| Benchmark | Qwen1.5-32B | Step 3 |
|---|---|---|
| LMArena Text | 1137 | 1351 |
| LMArena Creative Writing | 1083 | 1322 |
| LMArena Multi-Turn | 1140 | 1340 |
Frequently asked questions
Is Qwen1.5-32B better than Step 3?
Step 3 is the stronger model overall, scoring 39.9 to 30.0 on the Noometry Index.
Is Qwen1.5-32B or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 39.5 versus 31.1 in the Noometry coding category.
How many benchmarks do Qwen1.5-32B and Step 3 share?
15 benchmarks have published results for both models. Qwen1.5-32B has 21 scored results on Noometry and Step 3 has 17.