Model comparison
Gemma 3 4B vs Llama 3.2 90B
Gemma 3 4B and Llama 3.2 90B score almost the same on the Noometry Index (28.1 vs 27.5), so choose on price, context window or the category you care about most.
Last verified . 3 shared benchmarks.
Google
MetaSummary
- They share 3 benchmarks with published results for both. Gemma 3 4B scores higher in 1 category and Llama 3.2 90B in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 3.2 90B leads 21.7 to 11.8.
- The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 41% for Llama 3.2 90B.
Side by side
| Gemma 3 4B | Llama 3.2 90B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 28.1 | 27.5 |
| Released | 2025-03-12 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.04 | — |
| Output $ / M tokens | $0.08 | — |
| Results tracked | 22 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Gemma 3 4B: 35.9 (#215), Llama 3.2 90B: —
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| LMArena Coding | 1230 | — |
Agentic & Tool Use Llama 3.2 90B leads
Gemma 3 4B: 20.9 (#142), Llama 3.2 90B: 30.0 (#80)
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
| BALROG | — | 27.3% |
Reasoning Llama 3.2 90B leads
Gemma 3 4B: 13.2 (#335), Llama 3.2 90B: 21.7 (#217)
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| Epoch Capabilities Index | 116.02 | 125.5 |
| Kagi LLM Benchmark | 25.2% | — |
| Chess Puzzles | 0% | — |
| EnigmaEval | — | 0.4% |
| LMArena Hard Prompts | 1253 | — |
| DTBench | 50.9% | — |
| LMCA | 2.8% | — |
Math Gemma 3 4B leads
Gemma 3 4B: 16.8 (#292), Llama 3.2 90B: 11.1 (#308)
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.5% | 2.6% |
| LMArena Math | 1239 | — |
| MATH Level 5 | — | 39.4% |
Knowledge Llama 3.2 90B leads
Gemma 3 4B: 11.8 (#299), Llama 3.2 90B: 21.7 (#274)
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | 23.2% | 41% |
| Vectara Hallucination Rate | 6.4% | — |
| LMArena Expert | 1223 | — |
| MMLU | — | 80.3% |
Multimodal Not comparable
Gemma 3 4B: —, Llama 3.2 90B: 25.4 (#124)
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | — | 1000 |
| GeoBench | — | 52% |
Multilingual Not comparable
Gemma 3 4B: 42.5 (#194), Llama 3.2 90B: —
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| LMArena Non-English | 1273 | — |
| LMArena German | 1281 | — |
| LMArena Russian | 1294 | — |
Instruction Following Not comparable
Gemma 3 4B: 65.2 (#225), Llama 3.2 90B: —
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| LMArena Instruction Following | 1239 | — |
Long Context Not comparable
Gemma 3 4B: 38.7 (#194), Llama 3.2 90B: —
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1273 | — |
Writing & Preference Not comparable
Gemma 3 4B: 42.0 (#239), Llama 3.2 90B: —
| Benchmark | Gemma 3 4B | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1291 | — |
| LMArena Creative Writing | 1271 | — |
| EQ-Bench Creative Writing | 1068 | — |
| LMArena Multi-Turn | 1255 | — |
Frequently asked questions
Is Gemma 3 4B better than Llama 3.2 90B?
Gemma 3 4B and Llama 3.2 90B score almost the same on the Noometry Index (28.1 vs 27.5), so choose on price, context window or the category you care about most.
How many benchmarks do Gemma 3 4B and Llama 3.2 90B share?
3 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Llama 3.2 90B has 9.