Model comparison
Qwen2.5-Coder-32B vs Step 3.5 Flash
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 33.4 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 12 benchmarks with published results for both. Qwen2.5-Coder-32B scores higher in 0 categories and Step 3.5 Flash in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 3.5 Flash leads 42.4 to 22.6.
- Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Step 3.5 Flash accepts more context: 256K tokens versus 33K.
Side by side
| Qwen2.5-Coder-32B | Step 3.5 Flash | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 33.4 | 42.3 |
| Released | 2024-09-18 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 33K | 256K |
| Max output | 29K | 256K |
| Input $ / M tokens | $0.66 | $0.10 |
| Output $ / M tokens | $1 | $0.30 |
| Results tracked | 31 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
Qwen2.5-Coder-32B: 22.6 (#333), Step 3.5 Flash: 42.4 (#105)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1276 | 1436 |
| SWE-bench Verified (bash only) | 9% | — |
| Aider Polyglot | 16.4% | — |
| BigCodeBench Instruct | 49% | — |
| LiveBench Coding | 56.9% | — |
| BigCodeBench Complete | 58% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 77% | — |
Reasoning Too close to call
Qwen2.5-Coder-32B: 21.2 (#225), Step 3.5 Flash: 22.2 (#202)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1411 |
| NYT Connections (extended) | — | 28.4% |
| LiveBench Reasoning | 42.1% | — |
| LiveBench Data Analysis | 49.9% | — |
| Epoch Capabilities Index | 119.49 | — |
| HellaSwag | 83% | — |
| LiveBench | 46.2% | — |
| WinoGrande | 80.8% | — |
Math Step 3.5 Flash leads
Qwen2.5-Coder-32B: 33.3 (#204), Step 3.5 Flash: 42.6 (#84)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1251 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
| LiveBench Math | 46.6% | — |
| GSM8K | 93% | — |
Knowledge Step 3.5 Flash leads
Qwen2.5-Coder-32B: 33.4 (#203), Step 3.5 Flash: 39.6 (#132)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1221 | 1421 |
| ARC (AI2) Challenge | 70.5% | — |
| MMLU | 79.1% | — |
Multilingual Step 3.5 Flash leads
Qwen2.5-Coder-32B: 37.8 (#235), Step 3.5 Flash: 50.5 (#119)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1205 | 1385 |
| LMArena Chinese | 1222 | 1447 |
| LMArena Russian | 1228 | 1385 |
| LMArena French | — | 1421 |
| LMArena German | — | 1405 |
| LMArena Japanese | — | 1354 |
| LMArena Korean | — | 1352 |
| LMArena Spanish | — | 1419 |
Instruction Following Step 3.5 Flash leads
Qwen2.5-Coder-32B: 61.4 (#245), Step 3.5 Flash: 73.1 (#124)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1223 | 1385 |
| LiveBench Instruction Following | 58.7% | — |
Long Context Step 3.5 Flash leads
Qwen2.5-Coder-32B: 38.0 (#208), Step 3.5 Flash: 42.8 (#117)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1251 | 1402 |
Writing & Preference Step 3.5 Flash leads
Qwen2.5-Coder-32B: 41.6 (#240), Step 3.5 Flash: 58.8 (#113)
| Benchmark | Qwen2.5-Coder-32B | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1230 | 1403 |
| LMArena Creative Writing | 1174 | 1357 |
| LMArena Multi-Turn | 1222 | 1405 |
| LiveBench Language | 23.3% | — |
Frequently asked questions
Is Qwen2.5-Coder-32B better than Step 3.5 Flash?
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 33.4 on the Noometry Index.
Which is cheaper, Qwen2.5-Coder-32B or Step 3.5 Flash?
Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Qwen2.5-Coder-32B or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Step 3.5 Flash does, with 256K tokens against 33K.
How many benchmarks do Qwen2.5-Coder-32B and Step 3.5 Flash share?
12 benchmarks have published results for both models. Qwen2.5-Coder-32B has 31 scored results on Noometry and Step 3.5 Flash has 19.