Model comparison
Gemma 4 31B IT vs Llama 3.2 90B
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 27.5 on the Noometry Index.
Last verified . 4 shared benchmarks.
Google
MetaSummary
- They share 4 benchmarks with published results for both. Gemma 4 31B IT scores higher in 4 categories and Llama 3.2 90B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 11.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 2.6% for Llama 3.2 90B.
Side by side
| Gemma 4 31B IT | Llama 3.2 90B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 43.5 | 27.5 |
| Released | 2026-04-02 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.09 | — |
| Output $ / M tokens | $0.34 | — |
| Results tracked | 35 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Gemma 4 31B IT: 42.3 (#108), Llama 3.2 90B: —
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| WeirdML | 52.3% | — |
| LMArena Coding | 1459 | — |
| ALE-Bench | 925.5 | — |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Llama 3.2 90B: 30.0 (#80)
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| BALROG | — | 27.3% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), Llama 3.2 90B: 21.7 (#217)
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| Epoch Capabilities Index | 142.74 | 125.5 |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 5% | — |
| EnigmaEval | — | 0.4% |
| Thematic Generalization | 53% | — |
| LMArena Hard Prompts | 1448 | — |
| DTBench | 82.7% | — |
| LMCA | 39.3% | — |
| Surface Evolver Bench | 30.6% | — |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), Llama 3.2 90B: 11.1 (#308)
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 2.6% |
| LMArena Math | 1465 | — |
| MATH Level 5 | — | 39.4% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#151), Llama 3.2 90B: 21.7 (#274)
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | 75.8% | 41% |
| SimpleQA Verified | 10.4% | — |
| Vectara Hallucination Rate | 7.4% | — |
| LMArena Expert | 1465 | — |
| MMLU | — | 80.3% |
Multimodal Gemma 4 31B IT leads
Gemma 4 31B IT: 41.6 (#34), Llama 3.2 90B: 25.4 (#124)
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | 1277 | 1000 |
| GeoBench | — | 52% |
| LMArena Document | 1425 | — |
Multilingual Not comparable
Gemma 4 31B IT: 53.8 (#57), Llama 3.2 90B: —
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Non-English | 1431 | — |
| LMArena Chinese | 1476 | — |
| LMArena French | 1435 | — |
| LMArena Russian | 1460 | — |
| LMArena Spanish | 1444 | — |
Instruction Following Not comparable
Gemma 4 31B IT: 75.5 (#61), Llama 3.2 90B: —
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Instruction Following | 1433 | — |
Long Context Not comparable
Gemma 4 31B IT: 44.2 (#71), Llama 3.2 90B: —
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1446 | — |
Writing & Preference Not comparable
Gemma 4 31B IT: 60.5 (#96), Llama 3.2 90B: —
| Benchmark | Gemma 4 31B IT | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1443 | — |
| LMArena Creative Writing | 1415 | — |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
| LMArena Multi-Turn | 1452 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Llama 3.2 90B?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 27.5 on the Noometry Index.
How many benchmarks do Gemma 4 31B IT and Llama 3.2 90B share?
4 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Llama 3.2 90B has 9.