Model comparison
Qwen3 8B vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 33.7 on the Noometry Index.
Last verified . 0 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- The widest gap is in reasoning, where Step 3 leads 28.4 to 16.6.
Side by side
| Qwen3 8B | Step 3 | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 33.7 | 40.5 |
| Released | 2025-04 | — |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.18 | — |
| Output $ / M tokens | $0.70 | — |
| Results tracked | 11 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Qwen3 8B: 34.0 (#248), Step 3: 40.1 (#147)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| SciCode | 22.6% | — |
| LMArena Coding | — | 1367 |
Agentic & Tool Use Not comparable
Qwen3 8B: 30.2 (#78), Step 3: —
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 42.6% | — |
Reasoning Step 3 leads
Qwen3 8B: 16.6 (#303), Step 3: 28.4 (#105)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | — | 62.3% |
| CritPt | 0% | — |
| Chess Puzzles | 5% | — |
| LMArena Hard Prompts | — | 1355 |
| DTBench | 59.7% | — |
| LMCA | 8.8% | — |
| Epoch Capabilities Index | 136.17 | — |
Math Step 3 leads
Qwen3 8B: 34.9 (#191), Step 3: 37.6 (#148)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 56.1% | — |
| LMArena Math | — | 1366 |
Knowledge Too close to call
Qwen3 8B: 36.1 (#173), Step 3: 36.8 (#164)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| GPQA Diamond | 56.8% | — |
| Vectara Hallucination Rate | 4.8% | — |
| LMArena Expert | — | 1333 |
Multimodal Not comparable
Qwen3 8B: —, Step 3: 35.5 (#86)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Not comparable
Qwen3 8B: —, Step 3: 46.3 (#159)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| LMArena Non-English | — | 1327 |
| LMArena Chinese | — | 1397 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1269 |
| LMArena Russian | — | 1331 |
| LMArena Spanish | — | 1371 |
Instruction Following Not comparable
Qwen3 8B: —, Step 3: 70.4 (#164)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| LMArena Instruction Following | — | 1332 |
Long Context Step 3 leads
Qwen3 8B: 37.9 (#210), Step 3: 40.3 (#157)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| Fiction.LiveBench | 62.1% | — |
| LMArena Longer Query | — | 1326 |
Writing & Preference Not comparable
Qwen3 8B: —, Step 3: 54.3 (#151)
| Benchmark | Qwen3 8B | Step 3 |
|---|---|---|
| LMArena Text | — | 1350 |
| LMArena Creative Writing | — | 1321 |
| LMArena Multi-Turn | — | 1341 |
Frequently asked questions
Is Qwen3 8B better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 33.7 on the Noometry Index.
Is Qwen3 8B or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 34.0 in the Noometry coding category.
How many benchmarks do Qwen3 8B and Step 3 share?
0 benchmarks have published results for both models. Qwen3 8B has 11 scored results on Noometry and Step 3 has 17.