Model comparison
Qwen3-4B vs Qwen3.5 35B-A3B
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 31.9 on the Noometry Index.
Last verified . 5 shared benchmarks.
Alibaba (Qwen)Summary
- They share 5 benchmarks with published results for both. Qwen3-4B scores higher in 0 categories and Qwen3.5 35B-A3B in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 35B-A3B leads 47.8 to 33.0.
- The biggest single-benchmark swing is GPQA Diamond: 52.3% for Qwen3-4B and 83.5% for Qwen3.5 35B-A3B.
Side by side
| Qwen3-4B | Qwen3.5 35B-A3B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 31.9 | 42.0 |
| Released | 2025-04-29 | 2026-02-01 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $2 |
| Results tracked | 6 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Qwen3-4B: —, Qwen3.5 35B-A3B: 33.8 (#251)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena WebDev | — | 1254 |
| SciCode | — | 29.3% |
| LMArena Coding | — | 1410 |
Agentic & Tool Use Not comparable
Qwen3-4B: 27.6 (#100), Qwen3.5 35B-A3B: —
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 35.7% | — |
Reasoning Qwen3.5 35B-A3B leads
Qwen3-4B: 19.2 (#268), Qwen3.5 35B-A3B: 24.6 (#161)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| Chess Puzzles | 4% | 10% |
| CritPt | — | 0.6% |
| LMArena Hard Prompts | — | 1400 |
| DTBench | — | 80% |
| LMCA | — | 29.5% |
| Epoch Capabilities Index | — | 142.52 |
Math Qwen3.5 35B-A3B leads
Qwen3-4B: 29.7 (#240), Qwen3.5 35B-A3B: 39.9 (#97)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| MathArena Final-Answer Competitions | 38.5% | 56% |
| OTIS Mock AIME 2024-2025 | 52.2% | 70% |
| LMArena Math | — | 1404 |
Knowledge Qwen3.5 35B-A3B leads
Qwen3-4B: 33.0 (#208), Qwen3.5 35B-A3B: 47.8 (#79)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| GPQA Diamond | 52.3% | 83.5% |
| Vectara Hallucination Rate | 5.7% | 10.5% |
| LMArena Expert | — | 1408 |
Multilingual Not comparable
Qwen3-4B: —, Qwen3.5 35B-A3B: 50.0 (#127)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Non-English | — | 1378 |
| LMArena Chinese | — | 1457 |
| LMArena French | — | 1412 |
| LMArena German | — | 1367 |
| LMArena Japanese | — | 1325 |
| LMArena Korean | — | 1356 |
| LMArena Russian | — | 1376 |
| LMArena Spanish | — | 1392 |
Instruction Following Not comparable
Qwen3-4B: —, Qwen3.5 35B-A3B: 72.8 (#128)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Instruction Following | — | 1379 |
Long Context Not comparable
Qwen3-4B: —, Qwen3.5 35B-A3B: 42.4 (#127)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Longer Query | — | 1389 |
Writing & Preference Not comparable
Qwen3-4B: —, Qwen3.5 35B-A3B: 57.9 (#124)
| Benchmark | Qwen3-4B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Text | — | 1395 |
| LMArena Creative Writing | — | 1346 |
| LMArena Multi-Turn | — | 1390 |
Frequently asked questions
Is Qwen3-4B better than Qwen3.5 35B-A3B?
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 31.9 on the Noometry Index.
How many benchmarks do Qwen3-4B and Qwen3.5 35B-A3B share?
5 benchmarks have published results for both models. Qwen3-4B has 6 scored results on Noometry and Qwen3.5 35B-A3B has 28.