Model comparison
Qwen2.5 32B Instruct vs Qwen3-4B
Qwen3-4B is the stronger model overall, scoring 31.5 to 29.7 on the Noometry Index.
Last verified . 3 shared benchmarks.
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. Qwen2.5 32B Instruct scores higher in 0 categories and Qwen3-4B in 3 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3-4B leads 28.5 to 15.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 7.4% for Qwen2.5 32B Instruct and 52.2% for Qwen3-4B.
Side by side
| Qwen2.5 32B Instruct | Qwen3-4B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 29.7 | 31.5 |
| Released | 2024-09 | 2025-04-29 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.70 | — |
| Output $ / M tokens | $2.80 | — |
| Results tracked | 7 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Qwen2.5 32B Instruct: 38.1 (#165), Qwen3-4B: —
| Benchmark | Qwen2.5 32B Instruct | Qwen3-4B |
|---|---|---|
| BigCodeBench Instruct | 45% | — |
| BigCodeBench Complete | 52.3% | — |
Agentic & Tool Use Not comparable
Qwen2.5 32B Instruct: —, Qwen3-4B: 27.3 (#94)
| Benchmark | Qwen2.5 32B Instruct | Qwen3-4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 35.7% |
Reasoning Too close to call
Qwen2.5 32B Instruct: 18.5 (#262), Qwen3-4B: 18.5 (#264)
| Benchmark | Qwen2.5 32B Instruct | Qwen3-4B |
|---|---|---|
| Chess Puzzles | 0% | 4% |
| Epoch Capabilities Index | 128.52 | — |
Math Qwen3-4B leads
Qwen2.5 32B Instruct: 15.3 (#289), Qwen3-4B: 28.5 (#228)
| Benchmark | Qwen2.5 32B Instruct | Qwen3-4B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.4% | 52.2% |
| MathArena Final-Answer Competitions | — | 38.5% |
| MATH Level 5 | 56.1% | — |
Knowledge Qwen3-4B leads
Qwen2.5 32B Instruct: 24.9 (#259), Qwen3-4B: 33.0 (#198)
| Benchmark | Qwen2.5 32B Instruct | Qwen3-4B |
|---|---|---|
| GPQA Diamond | 46.1% | 52.3% |
| Vectara Hallucination Rate | — | 5.7% |
Frequently asked questions
Is Qwen2.5 32B Instruct better than Qwen3-4B?
Qwen3-4B is the stronger model overall, scoring 31.5 to 29.7 on the Noometry Index.
How many benchmarks do Qwen2.5 32B Instruct and Qwen3-4B share?
3 benchmarks have published results for both models. Qwen2.5 32B Instruct has 7 scored results on Noometry and Qwen3-4B has 6.