Model comparison
Gemma 2B vs Llama 3-8B
Gemma 2B is the stronger model overall, scoring 29.6 to 25.5 on the Noometry Index.
Last verified . 18 shared benchmarks.
Google
MetaSummary
- They share 18 benchmarks with published results for both. Gemma 2B scores higher in 2 categories and Llama 3-8B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 2B leads 30.0 to 8.8.
Side by side
| Gemma 2B | Llama 3-8B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 29.6 | 25.5 |
| Released | 2024-02-21 | 2024-04-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 23 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-8B leads
Gemma 2B: 29.4 (#305), Llama 3-8B: 31.0 (#289)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| LMArena Coding | 1010 | 1152 |
| HumanEval+ | 20.7% | 56.7% |
| MBPP+ | 34.1% | 54.8% |
| BigCodeBench Instruct | — | 31.9% |
| BigCodeBench Complete | — | 36.9% |
Reasoning Gemma 2B leads
Gemma 2B: 18.8 (#275), Llama 3-8B: 14.3 (#326)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| LMArena Hard Prompts | 989 | 1133 |
| Epoch Capabilities Index | 94.2 | 116.45 |
| WinoGrande | 65.4% | 75.7% |
| Chess Puzzles | — | 0% |
| DTBench | — | 43.9% |
| Adversarial NLI | — | 57.3% |
| BIG-Bench Hard | 35.2% | — |
| ForecastBench | — | 58.6 |
| HellaSwag | 71.4% | — |
| PIQA | 77.3% | — |
Math Gemma 2B leads
Gemma 2B: 30.0 (#239), Llama 3-8B: 8.8 (#323)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| LMArena Math | 1009 | 1151 |
| OTIS Mock AIME 2024-2025 | — | 1.9% |
| MATH Level 5 | — | 6.1% |
| GSM8K | 17.7% | — |
Knowledge Not comparable
Gemma 2B: —, Llama 3-8B: 7.8 (#308)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| ARC (AI2) Challenge | 42.1% | 82.8% |
| MMLU | 42.3% | 68.8% |
| TriviaQA | 53.2% | 67.7% |
| GPQA Diamond | — | 26.1% |
| LMArena Expert | — | 1113 |
| BoolQ | 69.4% | — |
| OpenBookQA | — | 82.6% |
Multilingual Llama 3-8B leads
Gemma 2B: 23.0 (#294), Llama 3-8B: 30.8 (#261)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| LMArena Non-English | 958 | 1098 |
| LMArena Chinese | 986 | 1076 |
| LMArena Russian | 937 | 1109 |
| LMArena French | — | 1159 |
| LMArena German | — | 1104 |
| LMArena Japanese | — | 967 |
| LMArena Korean | — | 1004 |
| LMArena Spanish | — | 1173 |
Instruction Following Llama 3-8B leads
Gemma 2B: 48.5 (#302), Llama 3-8B: 58.4 (#260)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| LMArena Instruction Following | 970 | 1127 |
Long Context Llama 3-8B leads
Gemma 2B: 29.9 (#291), Llama 3-8B: 34.2 (#251)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| LMArena Longer Query | 981 | 1128 |
Writing & Preference Llama 3-8B leads
Gemma 2B: 24.0 (#308), Llama 3-8B: 37.5 (#256)
| Benchmark | Gemma 2B | Llama 3-8B |
|---|---|---|
| LMArena Text | 1002 | 1166 |
| LMArena Creative Writing | 987 | 1150 |
| LMArena Multi-Turn | 945 | 1152 |
Frequently asked questions
Is Gemma 2B better than Llama 3-8B?
Gemma 2B is the stronger model overall, scoring 29.6 to 25.5 on the Noometry Index.
Is Gemma 2B or Llama 3-8B better for coding?
Llama 3-8B scores higher on coding benchmarks: 31.0 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2B and Llama 3-8B share?
18 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Llama 3-8B has 34.