Model comparison
Llama 4 Maverick vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.1 to 30.6 on the Noometry Index.
Last verified . 20 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Llama 4 Maverick scores higher in 1 category and Qwen2.5-Max in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5-Max leads 55.6 to 39.0.
- The biggest single-benchmark swing is Short-Story Creative Writing: 62% for Llama 4 Maverick and 72.9% for Qwen2.5-Max.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| Llama 4 Maverick | Qwen2.5-Max | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.6 | 40.1 |
| Released | 2025-04-05 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.19 | — |
| Output $ / M tokens | $0.65 | — |
| Results tracked | 54 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
Llama 4 Maverick: 26.3 (#318), Qwen2.5-Max: 41.2 (#118)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1302 | 1359 |
| SWE-bench Verified (bash only) | 21% | — |
| Aider Polyglot | 15.6% | — |
| SciCode | 33.1% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| LiveBench Coding | — | 64.4% |
| BigCodeBench Complete | 61.4% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Not comparable
Llama 4 Maverick: 27.9 (#86), Qwen2.5-Max: —
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
Reasoning Qwen2.5-Max leads
Llama 4 Maverick: 10.1 (#337), Qwen2.5-Max: 24.6 (#144)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1281 | 1360 |
| Epoch Capabilities Index | 132.2 | 132.54 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 27.7% | — |
| Kagi LLM Benchmark | 55.9% | — |
| NYT Connections (extended) | 8% | — |
| ARC-AGI-1 | 4.4% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.6% | — |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 61.9% | — |
| LiveBench Data Analysis | — | 67.9% |
| LMCA | 15.9% | — |
| ForecastBench | 57.5 | — |
| LiveBench | — | 62.3% |
Math Qwen2.5-Max leads
Llama 4 Maverick: 24.6 (#256), Qwen2.5-Max: 34.3 (#162)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1299 | 1369 |
| OTIS Mock AIME 2024-2025 | 20.6% | — |
| Omni-MATH | 42.2% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 73% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Qwen2.5-Max leads
Llama 4 Maverick: 33.3 (#196), Qwen2.5-Max: 34.8 (#181)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| Confabulations | 22.6% | 21.8% |
| LMArena Expert | 1259 | 1337 |
| GPQA Diamond | 67% | — |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Vectara Hallucination Rate | 8.2% | — |
| GPQA (HELM) | 65% | — |
Multimodal Not comparable
Llama 4 Maverick: 30.9 (#100), Qwen2.5-Max: —
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Vision | 1141 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Qwen2.5-Max leads
Llama 4 Maverick: 42.4 (#190), Qwen2.5-Max: 48.3 (#140)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1269 | 1352 |
| LMArena Chinese | 1277 | 1382 |
| LMArena French | 1258 | 1396 |
| LMArena German | 1291 | 1350 |
| LMArena Japanese | 1207 | 1299 |
| LMArena Korean | 1203 | 1304 |
| LMArena Russian | 1286 | 1353 |
| LMArena Spanish | 1293 | 1377 |
Instruction Following Too close to call
Llama 4 Maverick: 71.9 (#140), Qwen2.5-Max: 71.5 (#147)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1267 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 90.8% | — |
Long Context Qwen2.5-Max leads
Llama 4 Maverick: 31.4 (#274), Qwen2.5-Max: 41.3 (#137)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1280 | 1358 |
| Fiction.LiveBench | 46.2% | — |
Writing & Preference Qwen2.5-Max leads
Llama 4 Maverick: 39.0 (#247), Qwen2.5-Max: 55.6 (#141)
| Benchmark | Llama 4 Maverick | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1287 | 1367 |
| LMArena Creative Writing | 1266 | 1339 |
| Short-Story Creative Writing | 62% | 72.9% |
| LMArena Multi-Turn | 1289 | 1364 |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Llama 4 Maverick better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.1 to 30.6 on the Noometry Index.
Is Llama 4 Maverick or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.2 versus 26.3 in the Noometry coding category.
How many benchmarks do Llama 4 Maverick and Qwen2.5-Max share?
20 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Qwen2.5-Max has 27.