Model comparison
Gemma 4 26B A4B IT vs Llama 2-70B
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.5 to 24.4 on the Noometry Index.
Last verified . 18 shared benchmarks.
Google
MetaSummary
- They share 18 benchmarks with published results for both. Gemma 4 26B A4B IT scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 26B A4B IT leads 47.6 to 8.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 82.2% for Gemma 4 26B A4B IT and 0% for Llama 2-70B.
Side by side
| Gemma 4 26B A4B IT | Llama 2-70B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 43.5 | 24.4 |
| Released | 2026-04-02 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.0675 | — |
| Output $ / M tokens | $0.23 | — |
| Results tracked | 28 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 39.0 (#164), Llama 2-70B: 31.4 (#286)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1447 | 1079 |
| LMArena WebDev | 1359 | — |
| SciCode | 40% | — |
| WeirdML | 35.2% | — |
| ALE-Bench | 927.17 | — |
Reasoning Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 21.8 (#213), Llama 2-70B: 14.4 (#325)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1439 | 1073 |
| DTBench | 74.9% | 41.6% |
| Epoch Capabilities Index | 141.85 | 113.79 |
| CritPt | 0% | — |
| Chess Puzzles | 6% | — |
| LMCA | 29.7% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 47.6 (#67), Llama 2-70B: 8.1 (#326)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 82.2% | 0% |
| LMArena Math | 1470 | 1091 |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |
Knowledge Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 45.7 (#85), Llama 2-70B: 7.4 (#310)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 73.2% | 26.3% |
| LMArena Expert | 1447 | 1039 |
| Vectara Hallucination Rate | 5.2% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multimodal Not comparable
Gemma 4 26B A4B IT: 40.6 (#46), Llama 2-70B: —
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| LMArena Vision | 1260 | — |
Multilingual Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 53.1 (#71), Llama 2-70B: 27.7 (#274)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1421 | 1045 |
| LMArena Chinese | 1495 | 995 |
| LMArena French | 1460 | 1090 |
| LMArena Russian | 1434 | 1083 |
| LMArena Spanish | 1417 | 1143 |
| LMArena German | — | 1041 |
| LMArena Japanese | — | 927 |
| LMArena Korean | — | 964 |
Instruction Following Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 74.8 (#82), Llama 2-70B: 54.9 (#278)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1420 | 1071 |
Long Context Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 43.6 (#91), Llama 2-70B: 32.3 (#270)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1428 | 1062 |
Writing & Preference Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 58.6 (#115), Llama 2-70B: 32.3 (#279)
| Benchmark | Gemma 4 26B A4B IT | Llama 2-70B |
|---|---|---|
| LMArena Text | 1434 | 1115 |
| LMArena Creative Writing | 1402 | 1075 |
| LMArena Multi-Turn | 1441 | 1088 |
| EQ-Bench Creative Writing | 1305 | — |
Frequently asked questions
Is Gemma 4 26B A4B IT better than Llama 2-70B?
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.5 to 24.4 on the Noometry Index.
Is Gemma 4 26B A4B IT or Llama 2-70B better for coding?
Gemma 4 26B A4B IT scores higher on coding benchmarks: 39.0 versus 31.4 in the Noometry coding category.
How many benchmarks do Gemma 4 26B A4B IT and Llama 2-70B share?
18 benchmarks have published results for both models. Gemma 4 26B A4B IT has 28 scored results on Noometry and Llama 2-70B has 35.