Model comparison
Qwen3-30B-A3B vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 38.9 on the Noometry Index.
Last verified . 16 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 16 benchmarks with published results for both. Qwen3-30B-A3B scores higher in 4 categories and Step 3 in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Step 3 leads 40.3 to 31.0.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 54.9% for Qwen3-30B-A3B and 62.3% for Step 3.
Side by side
| Qwen3-30B-A3B | Step 3 | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 38.9 | 40.5 |
| Released | 2025-04-28 | — |
| Weights | Open | Open |
| Context window | 41K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.12 | — |
| Output $ / M tokens | $0.50 | — |
| Results tracked | 32 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Qwen3-30B-A3B: 37.5 (#194), Step 3: 40.1 (#147)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Coding | 1416 | 1367 |
| SciCode | 33.3% | — |
| WeirdML | 29.8% | — |
Agentic & Tool Use Not comparable
Qwen3-30B-A3B: 29.8 (#82), Step 3: —
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 41.4% | — |
Reasoning Step 3 leads
Qwen3-30B-A3B: 22.2 (#204), Step 3: 28.4 (#105)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 54.9% | 62.3% |
| LMArena Hard Prompts | 1398 | 1355 |
| CritPt | 0.3% | — |
| Chess Puzzles | 8% | — |
| DTBench | 69.3% | — |
| LMCA | 22.4% | — |
| Epoch Capabilities Index | 139.63 | — |
Math Too close to call
Qwen3-30B-A3B: 37.4 (#157), Step 3: 37.6 (#148)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Math | 1394 | 1366 |
| MathArena Final-Answer Competitions | 47.8% | — |
| OTIS Mock AIME 2024-2025 | 70.3% | — |
Knowledge Qwen3-30B-A3B leads
Qwen3-30B-A3B: 41.8 (#105), Step 3: 36.8 (#164)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Expert | 1396 | 1333 |
| GPQA Diamond | 70.1% | — |
| Confabulations | 12.3% | — |
Multimodal Not comparable
Qwen3-30B-A3B: —, Step 3: 35.5 (#86)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Qwen3-30B-A3B leads
Qwen3-30B-A3B: 49.5 (#132), Step 3: 46.3 (#159)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Non-English | 1372 | 1327 |
| LMArena Chinese | 1433 | 1397 |
| LMArena German | 1380 | 1371 |
| LMArena Korean | 1331 | 1269 |
| LMArena Russian | 1370 | 1331 |
| LMArena Spanish | 1404 | 1371 |
| LMArena French | 1418 | — |
| LMArena Japanese | 1337 | — |
Instruction Following Qwen3-30B-A3B leads
Qwen3-30B-A3B: 72.0 (#142), Step 3: 70.4 (#164)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1363 | 1332 |
Long Context Step 3 leads
Qwen3-30B-A3B: 31.0 (#283), Step 3: 40.3 (#157)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Longer Query | 1379 | 1326 |
| Fiction.LiveBench | 40.6% | — |
Writing & Preference Qwen3-30B-A3B leads
Qwen3-30B-A3B: 55.6 (#143), Step 3: 54.3 (#151)
| Benchmark | Qwen3-30B-A3B | Step 3 |
|---|---|---|
| LMArena Text | 1384 | 1350 |
| LMArena Creative Writing | 1317 | 1321 |
| LMArena Multi-Turn | 1378 | 1341 |
| Short-Story Creative Writing | 75.3% | — |
Frequently asked questions
Is Qwen3-30B-A3B better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 38.9 on the Noometry Index.
Is Qwen3-30B-A3B or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 37.5 in the Noometry coding category.
How many benchmarks do Qwen3-30B-A3B and Step 3 share?
16 benchmarks have published results for both models. Qwen3-30B-A3B has 32 scored results on Noometry and Step 3 has 17.