Model comparison
Llama 4 Maverick vs Llama 4 Scout
Llama 4 Maverick is the stronger model overall, scoring 30.9 to 27.7 on the Noometry Index. Llama 4 Scout costs 2.0× less per token, which makes it the better buy when Llama 4 Maverick's lead doesn't matter for your workload.
Last verified . 43 shared benchmarks.
MetaSummary
- They share 43 benchmarks with published results for both. Llama 4 Maverick scores higher in 9 categories and Llama 4 Scout in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Llama 4 Maverick leads 26.0 to 19.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 55.9% for Llama 4 Maverick and 36.9% for Llama 4 Scout.
- Llama 4 Scout is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
Side by side
| Llama 4 Maverick | Llama 4 Scout | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 30.9 | 27.7 |
| Released | 2025-04-05 | 2025-04-05 |
| Weights | Open | Open |
| Context window | 128K | 128K |
| Max output | 4K | 4K |
| Input $ / M tokens | $0.19 | $0.10 |
| Output $ / M tokens | $0.65 | $0.30 |
| Results tracked | 54 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 4 Maverick leads
Llama 4 Maverick: 26.6 (#324), Llama 4 Scout: 20.2 (#339)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| SWE-bench Verified (bash only) | 21% | 9.1% |
| SciCode | 33.1% | 17% |
| LMArena Coding | 1302 | 1286 |
| BigCodeBench Complete | 61.4% | 43.1% |
| Aider Polyglot | 15.6% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Llama 4 Maverick leads
Llama 4 Maverick: 28.2 (#91), Llama 4 Scout: 24.6 (#119)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | 28.1% |
Reasoning Too close to call
Llama 4 Maverick: 10.1 (#342), Llama 4 Scout: 9.1 (#345)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| ARC-AGI-2 | 0% | 0% |
| Kagi LLM Benchmark | 55.9% | 36.9% |
| ARC-AGI-1 | 4.4% | 0.5% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1281 | 1266 |
| DTBench | 61.9% | 57.9% |
| LMCA | 15.9% | 12% |
| Epoch Capabilities Index | 132.2 | 129.64 |
| ForecastBench | 57.5 | 57.5 |
| SimpleBench | 27.7% | — |
| NYT Connections (extended) | 8% | — |
| EnigmaEval | 0.6% | — |
Math Llama 4 Maverick leads
Llama 4 Maverick: 26.0 (#262), Llama 4 Scout: 19.6 (#286)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 20.6% | 7.8% |
| Omni-MATH | 42.2% | 37.3% |
| LMArena Math | 1299 | 1287 |
| MATH Level 5 | 73% | 62.3% |
| FrontierMath (Feb 2025 set) | 0.7% | 0% |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.4 (#204), Llama 4 Scout: 31.9 (#217)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| GPQA Diamond | 67% | 51.8% |
| MMLU-Pro | 81% | 74.2% |
| Vectara Hallucination Rate | 8.2% | 7.7% |
| GPQA (HELM) | 65% | 50.7% |
| LMArena Expert | 1259 | 1235 |
| Humanity's Last Exam | 5.7% | — |
| Confabulations | 22.6% | — |
Multimodal Too close to call
Llama 4 Maverick: 31.6 (#105), Llama 4 Scout: 32.2 (#102)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| LMArena Vision | 1142 | 1118 |
| SpatialViz-Bench | 31.8% | 34.2% |
| GeoBench | 52% | — |
Multilingual Llama 4 Maverick leads
Llama 4 Maverick: 42.2 (#195), Llama 4 Scout: 41.0 (#212)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| LMArena Non-English | 1269 | 1252 |
| LMArena Chinese | 1277 | 1255 |
| LMArena French | 1259 | 1282 |
| LMArena German | 1291 | 1272 |
| LMArena Japanese | 1207 | 1206 |
| LMArena Korean | 1203 | 1207 |
| LMArena Russian | 1286 | 1263 |
| LMArena Spanish | 1293 | 1278 |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.7 (#146), Llama 4 Scout: 65.8 (#217)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| IFEval | 90.8% | 81.8% |
| LMArena Instruction Following | 1267 | 1248 |
Long Context Llama 4 Maverick leads
Llama 4 Maverick: 31.4 (#279), Llama 4 Scout: 27.5 (#294)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| Fiction.LiveBench | 46.2% | 36% |
| LMArena Longer Query | 1280 | 1265 |
Writing & Preference Llama 4 Maverick leads
Llama 4 Maverick: 38.8 (#252), Llama 4 Scout: 37.0 (#261)
| Benchmark | Llama 4 Maverick | Llama 4 Scout |
|---|---|---|
| LMArena Text | 1287 | 1279 |
| LMArena Creative Writing | 1267 | 1249 |
| EQ-Bench Creative Writing | 860 | 783 |
| WildBench | 80% | 78% |
| LMArena Multi-Turn | 1289 | 1280 |
| Short-Story Creative Writing | 62% | — |
Frequently asked questions
Is Llama 4 Maverick better than Llama 4 Scout?
Llama 4 Maverick is the stronger model overall, scoring 30.9 to 27.7 on the Noometry Index. Llama 4 Scout costs 2.0× less per token, which makes it the better buy when Llama 4 Maverick's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Maverick or Llama 4 Scout?
Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.
Is Llama 4 Maverick or Llama 4 Scout better for coding?
Llama 4 Maverick scores higher on coding benchmarks: 26.6 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Both accept 128K tokens.
How many benchmarks do Llama 4 Maverick and Llama 4 Scout share?
43 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Llama 4 Scout has 43.