Model comparison
Gemma 2B vs Llama 3-70B
Gemma 2B and Llama 3-70B score almost the same on the Noometry Index (29.1 vs 28.5), so choose on price, context window or the category you care about most.
Last verified . 16 shared benchmarks.
Google
MetaSummary
- They share 16 benchmarks with published results for both. Gemma 2B scores higher in 2 categories and Llama 3-70B in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Llama 3-70B leads 43.0 to 24.1.
Side by side
| Gemma 2B | Llama 3-70B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 29.1 | 28.5 |
| Released | 2024-02-21 | 2024-04-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 23 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-70B leads
Gemma 2B: 28.9 (#300), Llama 3-70B: 35.3 (#215)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1010 | 1206 |
| HumanEval+ | 20.7% | 72% |
| MBPP+ | 34.1% | 69% |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
Agentic & Tool Use Not comparable
Gemma 2B: —, Llama 3-70B: 20.9 (#132)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| Cybench | — | 5% |
Reasoning Too close to call
Gemma 2B: 17.9 (#275), Llama 3-70B: 17.4 (#284)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 989 | 1195 |
| Epoch Capabilities Index | 93.57 | 122.91 |
| WinoGrande | 65.4% | 83.5% |
| Kagi LLM Benchmark | — | 35.1% |
| DTBench | — | 54.2% |
| BIG-Bench Hard | 35.2% | — |
| ForecastBench | — | 57.1 |
| HellaSwag | 71.4% | — |
| PIQA | 77.3% | — |
Math Gemma 2B leads
Gemma 2B: 27.6 (#237), Llama 3-70B: 12.0 (#299)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| LMArena Math | 1009 | 1218 |
| OTIS Mock AIME 2024-2025 | — | 4.3% |
| MATH Level 5 | — | 22.6% |
| GSM8K | 17.7% | — |
Knowledge Not comparable
Gemma 2B: —, Llama 3-70B: 20.8 (#272)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| MMLU | 42.3% | 79.3% |
| GPQA Diamond | — | 40.6% |
| LMArena Expert | — | 1149 |
| ARC (AI2) Challenge | 42.1% | — |
| BoolQ | 69.4% | — |
| TriviaQA | 53.2% | — |
Multilingual Llama 3-70B leads
Gemma 2B: 23.1 (#289), Llama 3-70B: 33.8 (#246)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 958 | 1142 |
| LMArena Chinese | 986 | 1114 |
| LMArena Russian | 937 | 1159 |
| LMArena French | — | 1231 |
| LMArena German | — | 1169 |
| LMArena Japanese | — | 1017 |
| LMArena Korean | — | 1017 |
| LMArena Spanish | — | 1241 |
Instruction Following Llama 3-70B leads
Gemma 2B: 48.8 (#297), Llama 3-70B: 62.8 (#233)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 971 | 1194 |
Long Context Llama 3-70B leads
Gemma 2B: 29.9 (#286), Llama 3-70B: 35.6 (#235)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 981 | 1174 |
Writing & Preference Llama 3-70B leads
Gemma 2B: 24.1 (#303), Llama 3-70B: 43.0 (#225)
| Benchmark | Gemma 2B | Llama 3-70B |
|---|---|---|
| LMArena Text | 1002 | 1221 |
| LMArena Creative Writing | 986 | 1210 |
| LMArena Multi-Turn | 945 | 1223 |
Frequently asked questions
Is Gemma 2B better than Llama 3-70B?
Gemma 2B and Llama 3-70B score almost the same on the Noometry Index (29.1 vs 28.5), so choose on price, context window or the category you care about most.
Is Gemma 2B or Llama 3-70B better for coding?
Llama 3-70B scores higher on coding benchmarks: 35.3 versus 28.9 in the Noometry coding category.
How many benchmarks do Gemma 2B and Llama 3-70B share?
16 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Llama 3-70B has 31.