Model comparison
Qwen3.8 Max vs Step 3
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 40.5 on the Noometry Index.
Last verified . 16 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 16 benchmarks with published results for both. Qwen3.8 Max scores higher in 9 categories and Step 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 37.6.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| Qwen3.8 Max | Step 3 | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 56.8 | 40.5 |
| Released | 2026-08-02 | — |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 131K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $6 | — |
| Results tracked | 39 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Qwen3.8 Max: 53.5 (#29), Step 3: 40.1 (#147)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Coding | 1502 | 1367 |
| DeepSWE | 57.5% | — |
| LMArena WebDev | 1674 | — |
| FrontierSWE | 17.8% | — |
| SciCode | 53.2% | — |
Agentic & Tool Use Not comparable
Qwen3.8 Max: 45.4 (#14), Step 3: —
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| APEX-Agents | 63.3% | — |
| τ²-bench Banking | 55.1% | — |
| GDP.pdf | 23.2% | — |
Reasoning Qwen3.8 Max leads
Qwen3.8 Max: 54.4 (#26), Step 3: 28.4 (#105)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1496 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| NYT Connections (extended) | 88.3% | — |
| CritPt | 20% | — |
| Chess Puzzles | 40% | — |
| Mystery Game Puzzles | 38% | — |
| DTBench | 92% | — |
| LMCA | 46.2% | — |
| Epoch Capabilities Index | 156.41 | — |
Math Qwen3.8 Max leads
Qwen3.8 Max: 73.2 (#20), Step 3: 37.6 (#148)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Math | 1499 | 1366 |
| FrontierMath (Tiers 1-3) | 74.7% | — |
| FrontierMath Tier 4 | 46.3% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 58% | — |
Knowledge Qwen3.8 Max leads
Qwen3.8 Max: 61.7 (#27), Step 3: 36.8 (#164)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Expert | 1507 | 1333 |
| GPQA Diamond | 92.7% | — |
| SimpleQA Verified | 47.3% | — |
Multimodal Qwen3.8 Max leads
Qwen3.8 Max: 37.2 (#75), Step 3: 35.5 (#86)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Vision | 1314 | 1177 |
| Furniture Assembly | 20% | — |
Multilingual Qwen3.8 Max leads
Qwen3.8 Max: 56.7 (#18), Step 3: 46.3 (#159)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Non-English | 1472 | 1327 |
| LMArena Chinese | 1538 | 1397 |
| LMArena German | 1483 | 1371 |
| LMArena Korean | 1461 | 1269 |
| LMArena Russian | 1481 | 1331 |
| LMArena Spanish | 1492 | 1371 |
| LMArena French | 1503 | — |
| LMArena Japanese | 1467 | — |
Instruction Following Qwen3.8 Max leads
Qwen3.8 Max: 77.6 (#17), Step 3: 70.4 (#164)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1479 | 1332 |
Long Context Qwen3.8 Max leads
Qwen3.8 Max: 45.6 (#31), Step 3: 40.3 (#157)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Longer Query | 1489 | 1326 |
Writing & Preference Qwen3.8 Max leads
Qwen3.8 Max: 67.1 (#30), Step 3: 54.3 (#151)
| Benchmark | Qwen3.8 Max | Step 3 |
|---|---|---|
| LMArena Text | 1483 | 1350 |
| LMArena Creative Writing | 1479 | 1321 |
| LMArena Multi-Turn | 1489 | 1341 |
Frequently asked questions
Is Qwen3.8 Max better than Step 3?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 40.5 on the Noometry Index.
Is Qwen3.8 Max or Step 3 better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 40.1 in the Noometry coding category.
How many benchmarks do Qwen3.8 Max and Step 3 share?
16 benchmarks have published results for both models. Qwen3.8 Max has 39 scored results on Noometry and Step 3 has 17.