Model comparison
Gemma 4 31B IT vs Llama 2-13B
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 29.6 on the Noometry Index.
Last verified . 17 shared benchmarks.
Google
MetaSummary
- They share 17 benchmarks with published results for both. Gemma 4 31B IT scores higher in 8 categories and Llama 2-13B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 4 31B IT leads 60.5 to 29.8.
- The biggest single-benchmark swing is DTBench: 82.7% for Gemma 4 31B IT and 42.2% for Llama 2-13B.
Side by side
| Gemma 4 31B IT | Llama 2-13B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 43.5 | 29.6 |
| Released | 2026-04-02 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.09 | — |
| Output $ / M tokens | $0.34 | — |
| Results tracked | 35 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 42.3 (#108), Llama 2-13B: 30.9 (#291)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1459 | 1062 |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| WeirdML | 52.3% | — |
| ALE-Bench | 925.5 | — |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), Llama 2-13B: 12.8 (#337)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| Chess Puzzles | 5% | 0% |
| LMArena Hard Prompts | 1448 | 1051 |
| DTBench | 82.7% | 42.2% |
| Epoch Capabilities Index | 142.74 | 106.17 |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| CritPt | 1.4% | — |
| Thematic Generalization | 53% | — |
| LMCA | 39.3% | — |
| Surface Evolver Bench | 30.6% | — |
| BIG-Bench Hard | — | 58.2% |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), Llama 2-13B: 31.1 (#229)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Math | 1465 | 1065 |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
| GSM8K | — | 36.9% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#151), Llama 2-13B: 28.1 (#249)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1465 | 1030 |
| GPQA Diamond | 75.8% | — |
| SimpleQA Verified | 10.4% | — |
| Vectara Hallucination Rate | 7.4% | — |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| MMLU | — | 55.6% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
Gemma 4 31B IT: 41.6 (#34), Llama 2-13B: —
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Vision | 1277 | — |
| LMArena Document | 1425 | — |
| ScienceQA | — | 55.8% |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), Llama 2-13B: 26.5 (#279)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1431 | 1024 |
| LMArena Chinese | 1476 | 1001 |
| LMArena French | 1435 | 1044 |
| LMArena Russian | 1460 | 1055 |
| LMArena Spanish | 1444 | 1087 |
| LMArena German | — | 1009 |
| LMArena Japanese | — | 894 |
| LMArena Korean | — | 953 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), Llama 2-13B: 53.3 (#287)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1045 |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), Llama 2-13B: 32.3 (#269)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1446 | 1064 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), Llama 2-13B: 29.8 (#289)
| Benchmark | Gemma 4 31B IT | Llama 2-13B |
|---|---|---|
| LMArena Text | 1443 | 1084 |
| LMArena Creative Writing | 1415 | 1047 |
| LMArena Multi-Turn | 1452 | 1050 |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Llama 2-13B?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 29.6 on the Noometry Index.
Is Gemma 4 31B IT or Llama 2-13B better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 30.9 in the Noometry coding category.
How many benchmarks do Gemma 4 31B IT and Llama 2-13B share?
17 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Llama 2-13B has 32.