Model comparison
Gemma 3 27B vs Llama 4 Maverick
Gemma 3 27B and Llama 4 Maverick score almost the same on the Noometry Index (30.8 vs 30.9), so choose on price, context window or the category you care about most.
Last verified . 35 shared benchmarks.
Google
MetaSummary
- They share 35 benchmarks with published results for both. Gemma 3 27B scores higher in 4 categories and Llama 4 Maverick in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 3 27B leads 52.5 to 38.8.
- The biggest single-benchmark swing is GPQA Diamond: 47.7% for Gemma 3 27B and 67% for Llama 4 Maverick.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
- Gemma 3 27B accepts more context: 131K tokens versus 128K.
Side by side
| Gemma 3 27B | Llama 4 Maverick | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 30.8 | 30.9 |
| Released | 2025-03-11 | 2025-04-05 |
| Weights | Open | Open |
| Context window | 131K | 128K |
| Max output | 8K | 4K |
| Input $ / M tokens | $0.08 | $0.19 |
| Output $ / M tokens | $0.16 | $0.65 |
| Results tracked | 43 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 4 Maverick leads
Gemma 3 27B: 22.5 (#334), Llama 4 Maverick: 26.6 (#324)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| Aider Polyglot | 4.9% | 15.6% |
| SciCode | 21.2% | 33.1% |
| LMArena Coding | 1322 | 1302 |
| SWE-bench Verified (bash only) | — | 21% |
| WeirdML | — | 24.5% |
| BigCodeBench Instruct | — | 49.7% |
| LiveBench Coding | 39.9% | — |
| BigCodeBench Complete | — | 61.4% |
| ALE-Bench | — | 172.97 |
Agentic & Tool Use Llama 4 Maverick leads
Gemma 3 27B: 25.1 (#110), Llama 4 Maverick: 28.2 (#91)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | 37.3% |
Reasoning Gemma 3 27B leads
Gemma 3 27B: 16.7 (#301), Llama 4 Maverick: 10.1 (#342)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| Kagi LLM Benchmark | 40.4% | 55.9% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1340 | 1281 |
| DTBench | 52.5% | 61.9% |
| LMCA | 12.3% | 15.9% |
| Epoch Capabilities Index | 130.04 | 132.2 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 27.7% |
| NYT Connections (extended) | — | 8% |
| ARC-AGI-1 | — | 4.4% |
| Chess Puzzles | 0% | — |
| EnigmaEval | — | 0.6% |
| LiveBench Reasoning | 43.8% | — |
| LiveBench Data Analysis | 51.5% | — |
| ForecastBench | — | 57.5 |
| LiveBench | 50% | — |
Math Too close to call
Gemma 3 27B: 25.9 (#265), Llama 4 Maverick: 26.0 (#262)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 20.6% |
| LMArena Math | 1312 | 1299 |
| MATH Level 5 | 74% | 73% |
| Omni-MATH | — | 42.2% |
| LiveBench Math | 55.4% | — |
| FrontierMath (Feb 2025 set) | — | 0.7% |
Knowledge Llama 4 Maverick leads
Gemma 3 27B: 25.5 (#261), Llama 4 Maverick: 33.4 (#204)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 47.7% | 67% |
| Confabulations | 40.3% | 22.6% |
| Vectara Hallucination Rate | 7.4% | 8.2% |
| LMArena Expert | 1304 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| MMLU-Pro | — | 81% |
| GPQA (HELM) | — | 65% |
Multimodal Gemma 3 27B leads
Gemma 3 27B: 32.6 (#100), Llama 4 Maverick: 31.6 (#105)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1164 | 1142 |
| GeoBench | 52% | 52% |
| SpatialViz-Bench | — | 31.8% |
Multilingual Gemma 3 27B leads
Gemma 3 27B: 46.9 (#155), Llama 4 Maverick: 42.2 (#195)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1334 | 1269 |
| LMArena Chinese | 1346 | 1277 |
| LMArena French | 1368 | 1259 |
| LMArena German | 1362 | 1291 |
| LMArena Japanese | 1287 | 1207 |
| LMArena Korean | 1308 | 1203 |
| LMArena Russian | 1349 | 1286 |
| LMArena Spanish | 1349 | 1293 |
Instruction Following Llama 4 Maverick leads
Gemma 3 27B: 70.6 (#160), Llama 4 Maverick: 71.7 (#146)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1321 | 1267 |
| LiveBench Instruction Following | 74.9% | — |
| IFEval | — | 90.8% |
Long Context Llama 4 Maverick leads
Gemma 3 27B: 27.6 (#293), Llama 4 Maverick: 31.4 (#279)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| Fiction.LiveBench | 33.3% | 46.2% |
| LMArena Longer Query | 1333 | 1280 |
Writing & Preference Gemma 3 27B leads
Gemma 3 27B: 52.5 (#168), Llama 4 Maverick: 38.8 (#252)
| Benchmark | Gemma 3 27B | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1358 | 1287 |
| LMArena Creative Writing | 1346 | 1267 |
| Short-Story Creative Writing | 79.9% | 62% |
| EQ-Bench Creative Writing | 1266 | 860 |
| LMArena Multi-Turn | 1345 | 1289 |
| WildBench | — | 80% |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Llama 4 Maverick?
Gemma 3 27B and Llama 4 Maverick score almost the same on the Noometry Index (30.8 vs 30.9), so choose on price, context window or the category you care about most.
Which is cheaper, Gemma 3 27B or Llama 4 Maverick?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.
Is Gemma 3 27B or Llama 4 Maverick better for coding?
Llama 4 Maverick scores higher on coding benchmarks: 26.6 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
Gemma 3 27B does, with 131K tokens against 128K.
How many benchmarks do Gemma 3 27B and Llama 4 Maverick share?
35 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Llama 4 Maverick has 54.