Model comparison
Gemma 3 4B vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 28.1 on the Noometry Index.
Last verified . 13 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Max leads 35.3 to 11.8.
- Gemma 3 4B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 4B | Qwen2.5-Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 28.1 | 40.7 |
| Released | 2025-03-12 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | 131K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.04 | — |
| Output $ / M tokens | $0.08 | — |
| Results tracked | 22 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
Gemma 3 4B: 35.9 (#215), Qwen2.5-Max: 41.8 (#117)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1230 | 1359 |
| LiveBench Coding | — | 64.4% |
Agentic & Tool Use Not comparable
Gemma 3 4B: 20.9 (#142), Qwen2.5-Max: —
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
Reasoning Qwen2.5-Max leads
Gemma 3 4B: 13.2 (#335), Qwen2.5-Max: 25.6 (#147)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1360 |
| Epoch Capabilities Index | 116.02 | 132.53 |
| Kagi LLM Benchmark | 25.2% | — |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 50.9% | — |
| LiveBench Data Analysis | — | 67.9% |
| LMCA | 2.8% | — |
| LiveBench | — | 62.3% |
Math Qwen2.5-Max leads
Gemma 3 4B: 16.8 (#292), Qwen2.5-Max: 36.9 (#162)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1239 | 1369 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
| LiveBench Math | — | 58.4% |
Knowledge Qwen2.5-Max leads
Gemma 3 4B: 11.8 (#299), Qwen2.5-Max: 35.3 (#186)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1223 | 1337 |
| GPQA Diamond | 23.2% | — |
| Confabulations | — | 21.8% |
| Vectara Hallucination Rate | 6.4% | — |
Multilingual Qwen2.5-Max leads
Gemma 3 4B: 42.5 (#194), Qwen2.5-Max: 48.1 (#146)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1273 | 1352 |
| LMArena German | 1281 | 1350 |
| LMArena Russian | 1294 | 1353 |
| LMArena Chinese | — | 1382 |
| LMArena French | — | 1396 |
| LMArena Japanese | — | 1300 |
| LMArena Korean | — | 1304 |
| LMArena Spanish | — | 1377 |
Instruction Following Qwen2.5-Max leads
Gemma 3 4B: 65.2 (#225), Qwen2.5-Max: 71.3 (#152)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1239 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
Long Context Qwen2.5-Max leads
Gemma 3 4B: 38.7 (#194), Qwen2.5-Max: 41.4 (#142)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1273 | 1358 |
Writing & Preference Qwen2.5-Max leads
Gemma 3 4B: 42.0 (#239), Qwen2.5-Max: 55.4 (#146)
| Benchmark | Gemma 3 4B | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1291 | 1367 |
| LMArena Creative Writing | 1271 | 1339 |
| LMArena Multi-Turn | 1255 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| EQ-Bench Creative Writing | 1068 | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Gemma 3 4B better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 28.1 on the Noometry Index.
Is Gemma 3 4B or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 35.9 in the Noometry coding category.
How many benchmarks do Gemma 3 4B and Qwen2.5-Max share?
13 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Qwen2.5-Max has 27.