Model comparison
Gemini 3.1 Pro Preview vs Llama 4 Maverick
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 30.9 on the Noometry Index. Llama 4 Maverick costs 15× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.
Last verified . 37 shared benchmarks.
Google
MetaSummary
- They share 37 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 10.1.
- The biggest single-benchmark swing is ARC-AGI-1: 98% for Gemini 3.1 Pro Preview and 4.4% for Llama 4 Maverick.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
- Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 128K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.1 Pro Preview | Llama 4 Maverick | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 56.7 | 30.9 |
| Released | 2026-02-19 | 2025-04-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 128K |
| Max output | 66K | 4K |
| Input $ / M tokens | $2 | $0.19 |
| Output $ / M tokens | $12 | $0.65 |
| Results tracked | 71 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 42.5 (#99), Llama 4 Maverick: 26.6 (#324)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| SciCode | 58.9% | 33.1% |
| WeirdML | 72.1% | 24.5% |
| LMArena Coding | 1484 | 1302 |
| ALE-Bench | 1,161 | 172.97 |
| SWE-bench Verified | 75.6% | — |
| DeepSWE | 11.7% | — |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| LMArena WebDev | 1447 | — |
| GSO | 22.6% | — |
| BigCodeBench Instruct | — | 49.7% |
| MirrorCode | 8.9% | — |
| BigCodeBench Complete | — | 61.4% |
| AlgoTune | 2.02 | — |
Agentic & Tool Use Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 37.7 (#34), Llama 4 Maverick: 28.2 (#91)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| Terminal-Bench | 80.2% | — |
| APEX-Agents | 35.3% | — |
| Berkeley Function Calling Leaderboard | — | 37.3% |
| τ²-bench Banking | 26% | — |
| DeepResearch Bench | 47.8% | — |
| PostTrainBench | 22% | — |
| BALROG | 57% | — |
| ExploitBench | 26.1% | — |
| GBAEval | 0.8% | — |
| GDP.pdf | 17% | — |
| LMArena Search | 1211 | — |
| METR Time Horizons | 77% | — |
| Vending-Bench 2 | 3,774 | — |
Reasoning Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 71.7 (#12), Llama 4 Maverick: 10.1 (#342)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| ARC-AGI-2 | 77.1% | 0% |
| SimpleBench | 79.6% | 27.7% |
| NYT Connections (extended) | 97.4% | 8% |
| ARC-AGI-1 | 98% | 4.4% |
| CritPt | 17.7% | 0% |
| EnigmaEval | 36.8% | 0.6% |
| LMArena Hard Prompts | 1485 | 1281 |
| DTBench | 97.1% | 61.9% |
| LMCA | 53.8% | 15.9% |
| Epoch Capabilities Index | 154.77 | 132.2 |
| ForecastBench | 59 | 57.5 |
| Kagi LLM Benchmark | — | 55.9% |
| Chess Puzzles | 55% | — |
| Thematic Generalization | 79.4% | — |
| EBR-Bench | 14.3% | — |
| Mystery Game Puzzles | 34% | — |
Math Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 62.1 (#34), Llama 4 Maverick: 26.0 (#262)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 95.6% | 20.6% |
| LMArena Math | 1485 | 1299 |
| FrontierMath (Feb 2025 set) | 36.9% | 0.7% |
| FrontierMath (Tiers 1-3) | 59.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 86.5% | — |
| ProofBench | 26% | — |
| Omni-MATH | — | 42.2% |
| MATH Level 5 | — | 73% |
| FrontierMath Tier 4 (v1) | 16.7% | — |
Knowledge Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 71.8 (#3), Llama 4 Maverick: 33.4 (#204)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 94.4% | 67% |
| Humanity's Last Exam | 46.4% | 5.7% |
| Vectara Hallucination Rate | 10.4% | 8.2% |
| LMArena Expert | 1485 | 1259 |
| SimpleQA Verified | 73.5% | — |
| MMLU-Pro | — | 81% |
| Confabulations | — | 22.6% |
| GPQA (HELM) | — | 65% |
Multimodal Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 37.9 (#69), Llama 4 Maverick: 31.6 (#105)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1296 | 1142 |
| GeoBench | — | 52% |
| Blueprint-Bench 2 | 26.5% | — |
| Furniture Assembly | 26.7% | — |
| LMArena Document | 1444 | — |
| SpatialViz-Bench | — | 31.8% |
Multilingual Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 57.0 (#12), Llama 4 Maverick: 42.2 (#195)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1477 | 1269 |
| LMArena Chinese | 1529 | 1277 |
| LMArena French | 1487 | 1259 |
| LMArena German | 1491 | 1291 |
| LMArena Japanese | 1493 | 1207 |
| LMArena Korean | 1455 | 1203 |
| LMArena Russian | 1498 | 1286 |
| LMArena Spanish | 1479 | 1293 |
Instruction Following Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 77.0 (#32), Llama 4 Maverick: 71.7 (#146)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1466 | 1267 |
| IFEval | — | 90.8% |
Long Context Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 47.4 (#18), Llama 4 Maverick: 31.4 (#279)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1483 | 1280 |
| Fiction.LiveBench | — | 46.2% |
| CL-bench | 20.8% | — |
| CL-bench Life | 16.9% | — |
Writing & Preference Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 66.1 (#37), Llama 4 Maverick: 38.8 (#252)
| Benchmark | Gemini 3.1 Pro Preview | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1481 | 1287 |
| LMArena Creative Writing | 1482 | 1267 |
| EQ-Bench Creative Writing | 1491 | 860 |
| LMArena Multi-Turn | 1488 | 1289 |
| Short-Story Creative Writing | — | 62% |
| WildBench | — | 80% |
| EQ-Bench 4 | 1142 | — |
Frequently asked questions
Is Gemini 3.1 Pro Preview better than Llama 4 Maverick?
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 30.9 on the Noometry Index. Llama 4 Maverick costs 15× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.1 Pro Preview or Llama 4 Maverick?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.
Is Gemini 3.1 Pro Preview or Llama 4 Maverick better for coding?
Gemini 3.1 Pro Preview scores higher on coding benchmarks: 42.5 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.1 Pro Preview does, with 1.05M tokens against 128K.
How many benchmarks do Gemini 3.1 Pro Preview and Llama 4 Maverick share?
37 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Llama 4 Maverick has 54.