Model comparison
Gemma 4 26B A4B IT vs Llama 3.2 90B
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.1 to 27.1 on the Noometry Index.
Last verified . 4 shared benchmarks.
Google
MetaSummary
- They share 4 benchmarks with published results for both. Gemma 4 26B A4B IT scores higher in 4 categories and Llama 3.2 90B in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 26B A4B IT leads 45.7 to 10.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 82.2% for Gemma 4 26B A4B IT and 2.6% for Llama 3.2 90B.
Side by side
| Gemma 4 26B A4B IT | Llama 3.2 90B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 43.1 | 27.1 |
| Released | 2026-04-02 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.0765 | — |
| Output $ / M tokens | $0.26 | — |
| Results tracked | 28 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Gemma 4 26B A4B IT: 38.8 (#156), Llama 3.2 90B: —
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| LMArena WebDev | 1359 | — |
| SciCode | 40% | — |
| WeirdML | 35.2% | — |
| LMArena Coding | 1446 | — |
| ALE-Bench | 927.17 | — |
Agentic & Tool Use Not comparable
Gemma 4 26B A4B IT: —, Llama 3.2 90B: 29.7 (#73)
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| BALROG | — | 27.3% |
Reasoning Too close to call
Gemma 4 26B A4B IT: 21.1 (#205), Llama 3.2 90B: 20.9 (#207)
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| Epoch Capabilities Index | 141.85 | 125.5 |
| CritPt | 0% | — |
| Chess Puzzles | 6% | — |
| EnigmaEval | — | 0.4% |
| LMArena Hard Prompts | 1438 | — |
| DTBench | 74.9% | — |
| LMCA | 29.7% | — |
Math Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 45.7 (#61), Llama 3.2 90B: 10.4 (#302)
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 82.2% | 2.6% |
| LMArena Math | 1469 | — |
| MATH Level 5 | — | 39.4% |
Knowledge Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 45.6 (#79), Llama 3.2 90B: 21.7 (#269)
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | 73.2% | 41% |
| Vectara Hallucination Rate | 5.2% | — |
| LMArena Expert | 1447 | — |
| MMLU | — | 80.3% |
Multimodal Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 39.9 (#43), Llama 3.2 90B: 24.9 (#119)
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | 1260 | 1000 |
| GeoBench | — | 52% |
Multilingual Not comparable
Gemma 4 26B A4B IT: 53.3 (#68), Llama 3.2 90B: —
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Non-English | 1422 | — |
| LMArena Chinese | 1495 | — |
| LMArena French | 1459 | — |
| LMArena Russian | 1434 | — |
| LMArena Spanish | 1415 | — |
Instruction Following Not comparable
Gemma 4 26B A4B IT: 75.0 (#78), Llama 3.2 90B: —
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Instruction Following | 1420 | — |
Long Context Not comparable
Gemma 4 26B A4B IT: 43.6 (#87), Llama 3.2 90B: —
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1427 | — |
Writing & Preference Not comparable
Gemma 4 26B A4B IT: 58.8 (#110), Llama 3.2 90B: —
| Benchmark | Gemma 4 26B A4B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1434 | — |
| LMArena Creative Writing | 1401 | — |
| EQ-Bench Creative Writing | 1305 | — |
| LMArena Multi-Turn | 1442 | — |
Frequently asked questions
Is Gemma 4 26B A4B IT better than Llama 3.2 90B?
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.1 to 27.1 on the Noometry Index.
How many benchmarks do Gemma 4 26B A4B IT and Llama 3.2 90B share?
4 benchmarks have published results for both models. Gemma 4 26B A4B IT has 28 scored results on Noometry and Llama 3.2 90B has 9.