Model comparison
Qwen2.5 72B Instruct vs Step 3.5 Flash
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 31.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 17 benchmarks with published results for both. Qwen2.5 72B Instruct scores higher in 1 category and Step 3.5 Flash in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3.5 Flash leads 42.6 to 19.3.
- Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Step 3.5 Flash accepts more context: 256K tokens versus 131K.
Side by side
| Qwen2.5 72B Instruct | Step 3.5 Flash | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 31.9 | 42.3 |
| Released | 2024-09 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 131K | 256K |
| Max output | 8K | 256K |
| Input $ / M tokens | $1.40 | $0.10 |
| Output $ / M tokens | $5.60 | $0.30 |
| Results tracked | 43 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
Qwen2.5 72B Instruct: 33.2 (#260), Step 3.5 Flash: 42.4 (#105)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1292 | 1436 |
| WeirdML | 16% | — |
| BigCodeBench Instruct | 45.8% | — |
| BigCodeBench Complete | 55.9% | — |
Agentic & Tool Use Not comparable
Qwen2.5 72B Instruct: 22.1 (#133), Step 3.5 Flash: —
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| TheAgentCompany | 5.7% | — |
| BALROG | 16.2% | — |
| METR Time Horizons | 35.8% | — |
Reasoning Too close to call
Qwen2.5 72B Instruct: 22.3 (#199), Step 3.5 Flash: 22.2 (#202)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1271 | 1411 |
| NYT Connections (extended) | — | 28.4% |
| DTBench | 62.9% | — |
| LMCA | 13.4% | — |
| BIG-Bench Hard | 79.8% | — |
| Epoch Capabilities Index | 129 | — |
| ForecastBench | 57.5 | — |
| HellaSwag | 84.8% | — |
| PIQA | 82.6% | — |
| WinoGrande | 82.3% | — |
Math Step 3.5 Flash leads
Qwen2.5 72B Instruct: 19.3 (#287), Step 3.5 Flash: 42.6 (#84)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1283 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
| OTIS Mock AIME 2024-2025 | 8.1% | — |
| Omni-MATH | 33% | — |
| MATH Level 5 | 63.2% | — |
Knowledge Step 3.5 Flash leads
Qwen2.5 72B Instruct: 27.0 (#253), Step 3.5 Flash: 39.6 (#132)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1245 | 1421 |
| GPQA Diamond | 49.1% | — |
| MMLU-Pro | 63.1% | — |
| Confabulations | 19.1% | — |
| GPQA (HELM) | 42.6% | — |
| ARC (AI2) Challenge | 94.5% | — |
| MMLU | 85.3% | — |
| TriviaQA | 71.9% | — |
Multilingual Step 3.5 Flash leads
Qwen2.5 72B Instruct: 41.0 (#213), Step 3.5 Flash: 50.5 (#119)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1252 | 1385 |
| LMArena Chinese | 1272 | 1447 |
| LMArena French | 1280 | 1421 |
| LMArena German | 1234 | 1405 |
| LMArena Japanese | 1180 | 1354 |
| LMArena Korean | 1188 | 1352 |
| LMArena Russian | 1264 | 1385 |
| LMArena Spanish | 1256 | 1419 |
Instruction Following Step 3.5 Flash leads
Qwen2.5 72B Instruct: 65.5 (#221), Step 3.5 Flash: 73.1 (#124)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1254 | 1385 |
| IFEval | 80.6% | — |
Long Context Step 3.5 Flash leads
Qwen2.5 72B Instruct: 38.9 (#188), Step 3.5 Flash: 42.8 (#117)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1282 | 1402 |
Writing & Preference Step 3.5 Flash leads
Qwen2.5 72B Instruct: 46.7 (#215), Step 3.5 Flash: 58.8 (#113)
| Benchmark | Qwen2.5 72B Instruct | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1269 | 1403 |
| LMArena Creative Writing | 1221 | 1357 |
| LMArena Multi-Turn | 1272 | 1405 |
| WildBench | 80.2% | — |
Frequently asked questions
Is Qwen2.5 72B Instruct better than Step 3.5 Flash?
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 31.9 on the Noometry Index.
Which is cheaper, Qwen2.5 72B Instruct 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 72B Instruct lists at $1.40 and $5.60.
Is Qwen2.5 72B Instruct or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
Step 3.5 Flash does, with 256K tokens against 131K.
How many benchmarks do Qwen2.5 72B Instruct and Step 3.5 Flash share?
17 benchmarks have published results for both models. Qwen2.5 72B Instruct has 43 scored results on Noometry and Step 3.5 Flash has 19.