Model comparison
Gemma 3 4B vs Qwen-14B
Qwen-14B is the stronger model overall, scoring 31.4 to 28.1 on the Noometry Index.
Last verified . 10 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 10 benchmarks with published results for both. Gemma 3 4B scores higher in 5 categories and Qwen-14B in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 3 4B leads 42.5 to 27.5.
Side by side
| Gemma 3 4B | Qwen-14B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 28.1 | 31.4 |
| Released | 2025-03-12 | 2023-09-24 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.04 | — |
| Output $ / M tokens | $0.08 | — |
| Results tracked | 22 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 4B leads
Gemma 3 4B: 35.9 (#215), Qwen-14B: 31.2 (#288)
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| LMArena Coding | 1230 | 1071 |
Agentic & Tool Use Not comparable
Gemma 3 4B: 20.9 (#142), Qwen-14B: —
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
Reasoning Qwen-14B leads
Gemma 3 4B: 13.2 (#335), Qwen-14B: 19.6 (#257)
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1027 |
| Epoch Capabilities Index | 116.02 | 113.03 |
| Kagi LLM Benchmark | 25.2% | — |
| Chess Puzzles | 0% | — |
| DTBench | 50.9% | — |
| LMCA | 2.8% | — |
| BIG-Bench Hard | — | 55% |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
Math Qwen-14B leads
Gemma 3 4B: 16.8 (#292), Qwen-14B: 31.2 (#227)
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| LMArena Math | 1239 | 1068 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
| GSM8K | — | 61.3% |
Knowledge Not comparable
Gemma 3 4B: 11.8 (#299), Qwen-14B: —
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| GPQA Diamond | 23.2% | — |
| Vectara Hallucination Rate | 6.4% | — |
| LMArena Expert | 1223 | — |
| ARC (AI2) Challenge | — | 84.4% |
| BoolQ | — | 86.2% |
| MMLU | — | 66.3% |
Multilingual Gemma 3 4B leads
Gemma 3 4B: 42.5 (#194), Qwen-14B: 27.5 (#275)
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| LMArena Non-English | 1273 | 1041 |
| LMArena Chinese | — | 1077 |
| LMArena German | 1281 | — |
| LMArena Russian | 1294 | — |
Instruction Following Gemma 3 4B leads
Gemma 3 4B: 65.2 (#225), Qwen-14B: 52.4 (#289)
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | 1239 | 1031 |
Long Context Gemma 3 4B leads
Gemma 3 4B: 38.7 (#194), Qwen-14B: 31.3 (#280)
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| LMArena Longer Query | 1273 | 1028 |
Writing & Preference Gemma 3 4B leads
Gemma 3 4B: 42.0 (#239), Qwen-14B: 27.6 (#299)
| Benchmark | Gemma 3 4B | Qwen-14B |
|---|---|---|
| LMArena Text | 1291 | 1051 |
| LMArena Creative Writing | 1271 | 1028 |
| LMArena Multi-Turn | 1255 | 1022 |
| EQ-Bench Creative Writing | 1068 | — |
Frequently asked questions
Is Gemma 3 4B better than Qwen-14B?
Qwen-14B is the stronger model overall, scoring 31.4 to 28.1 on the Noometry Index.
Is Gemma 3 4B or Qwen-14B better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 31.2 in the Noometry coding category.
How many benchmarks do Gemma 3 4B and Qwen-14B share?
10 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Qwen-14B has 18.