Model comparison
Qwen3.8 27B vs Step 3
Qwen3.8 27B is the stronger model overall, scoring 46.0 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 27B scores higher in 8 categories and Step 3 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 28.4.
Side by side
| Qwen3.8 27B | Step 3 | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 46.0 | 40.5 |
| Released | 2026-08-14 | — |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.99 | — |
| Output $ / M tokens | $1.49 | — |
| Results tracked | 31 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Qwen3.8 27B: 50.5 (#44), Step 3: 40.1 (#147)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Coding | 1482 | 1367 |
| LMArena WebDev | 1593 | — |
| SciCode | 46.6% | — |
Agentic & Tool Use Not comparable
Qwen3.8 27B: 32.9 (#57), Step 3: —
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| APEX-Agents | 47.5% | — |
Reasoning Qwen3.8 27B leads
Qwen3.8 27B: 41.0 (#54), Step 3: 28.4 (#105)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1460 | 1355 |
| ARC-AGI-2 | 42.4% | — |
| Kagi LLM Benchmark | — | 62.3% |
| NYT Connections (extended) | 54.5% | — |
| ARC-AGI-1 | 87.5% | — |
| CritPt | 5.4% | — |
| DTBench | 88% | — |
| LMCA | 41.4% | — |
| Surface Evolver Bench | 45% | — |
| Epoch Capabilities Index | 149.38 | — |
Math Too close to call
Qwen3.8 27B: 37.1 (#161), Step 3: 37.6 (#148)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Math | 1456 | 1366 |
| ProofBench | 16% | — |
Knowledge Qwen3.8 27B leads
Qwen3.8 27B: 41.6 (#109), Step 3: 36.8 (#164)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Expert | 1482 | 1333 |
Multimodal Qwen3.8 27B leads
Qwen3.8 27B: 41.3 (#37), Step 3: 35.5 (#86)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Vision | 1271 | 1177 |
Multilingual Qwen3.8 27B leads
Qwen3.8 27B: 53.7 (#60), Step 3: 46.3 (#159)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Non-English | 1430 | 1327 |
| LMArena Chinese | 1504 | 1397 |
| LMArena German | 1438 | 1371 |
| LMArena Korean | 1393 | 1269 |
| LMArena Russian | 1415 | 1331 |
| LMArena Spanish | 1448 | 1371 |
| LMArena French | 1465 | — |
| LMArena Japanese | 1384 | — |
Instruction Following Qwen3.8 27B leads
Qwen3.8 27B: 75.8 (#53), Step 3: 70.4 (#164)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1439 | 1332 |
Long Context Qwen3.8 27B leads
Qwen3.8 27B: 44.3 (#70), Step 3: 40.3 (#157)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Longer Query | 1450 | 1326 |
Writing & Preference Qwen3.8 27B leads
Qwen3.8 27B: 65.8 (#43), Step 3: 54.3 (#151)
| Benchmark | Qwen3.8 27B | Step 3 |
|---|---|---|
| LMArena Text | 1441 | 1350 |
| LMArena Creative Writing | 1384 | 1321 |
| LMArena Multi-Turn | 1441 | 1341 |
| EQ-Bench Creative Writing | 1671 | — |
Frequently asked questions
Is Qwen3.8 27B better than Step 3?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 40.5 on the Noometry Index.
Is Qwen3.8 27B or Step 3 better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 40.1 in the Noometry coding category.
How many benchmarks do Qwen3.8 27B and Step 3 share?
16 benchmarks have published results for both models. Qwen3.8 27B has 31 scored results on Noometry and Step 3 has 17.