Model comparison
Gemma 2B vs Gemma 3 4B
Gemma 2B is the stronger model overall, scoring 29.6 to 28.1 on the Noometry Index.
Last verified . 11 shared benchmarks.
GoogleSummary
- They share 11 benchmarks with published results for both. Gemma 2B scores higher in 2 categories and Gemma 3 4B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 3 4B leads 42.5 to 23.0.
Side by side
| Gemma 2B | Gemma 3 4B | |
|---|---|---|
| Provider | ||
| Noometry Index | 29.6 | 28.1 |
| Released | 2024-02-21 | 2025-03-12 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.04 |
| Output $ / M tokens | — | $0.08 |
| Results tracked | 23 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 4B leads
Gemma 2B: 29.4 (#305), Gemma 3 4B: 35.9 (#215)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| LMArena Coding | 1010 | 1230 |
| HumanEval+ | 20.7% | — |
| MBPP+ | 34.1% | — |
Agentic & Tool Use Not comparable
Gemma 2B: —, Gemma 3 4B: 20.9 (#142)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 19.6% |
Reasoning Gemma 2B leads
Gemma 2B: 18.8 (#275), Gemma 3 4B: 13.2 (#335)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| LMArena Hard Prompts | 989 | 1253 |
| Epoch Capabilities Index | 94.2 | 116.02 |
| Kagi LLM Benchmark | — | 25.2% |
| Chess Puzzles | — | 0% |
| DTBench | — | 50.9% |
| LMCA | — | 2.8% |
| BIG-Bench Hard | 35.2% | — |
| HellaSwag | 71.4% | — |
| PIQA | 77.3% | — |
| WinoGrande | 65.4% | — |
Math Gemma 2B leads
Gemma 2B: 30.0 (#239), Gemma 3 4B: 16.8 (#292)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| LMArena Math | 1009 | 1239 |
| OTIS Mock AIME 2024-2025 | — | 7.5% |
| GSM8K | 17.7% | — |
Knowledge Not comparable
Gemma 2B: —, Gemma 3 4B: 11.8 (#299)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| GPQA Diamond | — | 23.2% |
| Vectara Hallucination Rate | — | 6.4% |
| LMArena Expert | — | 1223 |
| ARC (AI2) Challenge | 42.1% | — |
| BoolQ | 69.4% | — |
| MMLU | 42.3% | — |
| TriviaQA | 53.2% | — |
Multilingual Gemma 3 4B leads
Gemma 2B: 23.0 (#294), Gemma 3 4B: 42.5 (#194)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| LMArena Non-English | 958 | 1273 |
| LMArena Russian | 937 | 1294 |
| LMArena Chinese | 986 | — |
| LMArena German | — | 1281 |
Instruction Following Gemma 3 4B leads
Gemma 2B: 48.5 (#302), Gemma 3 4B: 65.2 (#225)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| LMArena Instruction Following | 970 | 1239 |
Long Context Gemma 3 4B leads
Gemma 2B: 29.9 (#291), Gemma 3 4B: 38.7 (#194)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| LMArena Longer Query | 981 | 1273 |
Writing & Preference Gemma 3 4B leads
Gemma 2B: 24.0 (#308), Gemma 3 4B: 42.0 (#239)
| Benchmark | Gemma 2B | Gemma 3 4B |
|---|---|---|
| LMArena Text | 1002 | 1291 |
| LMArena Creative Writing | 987 | 1271 |
| LMArena Multi-Turn | 945 | 1255 |
| EQ-Bench Creative Writing | — | 1068 |
Frequently asked questions
Is Gemma 2B better than Gemma 3 4B?
Gemma 2B is the stronger model overall, scoring 29.6 to 28.1 on the Noometry Index.
Is Gemma 2B or Gemma 3 4B better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2B and Gemma 3 4B share?
11 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Gemma 3 4B has 22.