Model comparison
Llama 4 Maverick vs Qwen Max
Qwen Max is the stronger model overall, scoring 34.7 to 30.9 on the Noometry Index. Llama 4 Maverick costs 9.2× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Llama 4 Maverick scores higher in 4 categories and Qwen Max in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen Max leads 25.1 to 10.1.
- The biggest single-benchmark swing is Fiction.LiveBench: 46.2% for Llama 4 Maverick and 66.7% for Qwen Max.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $1.60 / $6.40 for Qwen Max.
- Llama 4 Maverick accepts more context: 128K tokens versus 33K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| Llama 4 Maverick | Qwen Max | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.9 | 34.7 |
| Released | 2025-04-05 | 2024-04-03 |
| Weights | Open | Proprietary |
| Context window | 128K | 33K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.19 | $1.60 |
| Output $ / M tokens | $0.65 | $6.40 |
| Results tracked | 54 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen Max leads
Llama 4 Maverick: 26.6 (#324), Qwen Max: 30.7 (#292)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| Aider Polyglot | 15.6% | 21.8% |
| LMArena Coding | 1302 | 1288 |
| SWE-bench Verified (bash only) | 21% | — |
| SciCode | 33.1% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| BigCodeBench Complete | 61.4% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Not comparable
Llama 4 Maverick: 28.2 (#91), Qwen Max: —
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
Reasoning Qwen Max leads
Llama 4 Maverick: 10.1 (#342), Qwen Max: 25.1 (#151)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| LMArena Hard Prompts | 1281 | 1269 |
| 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% | — |
| DTBench | 61.9% | — |
| LMCA | 15.9% | — |
| Epoch Capabilities Index | 132.2 | — |
| ForecastBench | 57.5 | — |
Math Llama 4 Maverick leads
Llama 4 Maverick: 26.0 (#262), Qwen Max: 22.3 (#276)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 20.6% | 16.1% |
| LMArena Math | 1299 | 1275 |
| MATH Level 5 | 73% | 67.2% |
| FrontierMath (Feb 2025 set) | 0.7% | 1% |
| Omni-MATH | 42.2% | — |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.4 (#204), Qwen Max: 30.3 (#228)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| GPQA Diamond | 67% | 56.1% |
| LMArena Expert | 1259 | 1248 |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Confabulations | 22.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
| GPQA (HELM) | 65% | — |
Multimodal Not comparable
Llama 4 Maverick: 31.6 (#105), Qwen Max: —
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| LMArena Vision | 1142 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Too close to call
Llama 4 Maverick: 42.2 (#195), Qwen Max: 41.8 (#202)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| LMArena Non-English | 1269 | 1263 |
| LMArena Chinese | 1277 | 1254 |
| LMArena French | 1259 | 1330 |
| LMArena German | 1291 | 1254 |
| LMArena Japanese | 1207 | 1205 |
| LMArena Korean | 1203 | 1142 |
| LMArena Russian | 1286 | 1274 |
| LMArena Spanish | 1293 | 1290 |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.7 (#146), Qwen Max: 66.5 (#208)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| LMArena Instruction Following | 1267 | 1262 |
| IFEval | 90.8% | — |
Long Context Qwen Max leads
Llama 4 Maverick: 31.4 (#279), Qwen Max: 39.4 (#180)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| Fiction.LiveBench | 46.2% | 66.7% |
| LMArena Longer Query | 1280 | 1288 |
Writing & Preference Qwen Max leads
Llama 4 Maverick: 38.8 (#252), Qwen Max: 47.8 (#205)
| Benchmark | Llama 4 Maverick | Qwen Max |
|---|---|---|
| LMArena Text | 1287 | 1282 |
| LMArena Creative Writing | 1267 | 1248 |
| LMArena Multi-Turn | 1289 | 1277 |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
Frequently asked questions
Is Llama 4 Maverick better than Qwen Max?
Qwen Max is the stronger model overall, scoring 34.7 to 30.9 on the Noometry Index. Llama 4 Maverick costs 9.2× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Maverick or Qwen Max?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Qwen Max lists at $1.60 and $6.40.
Is Llama 4 Maverick or Qwen Max better for coding?
Qwen Max scores higher on coding benchmarks: 30.7 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Llama 4 Maverick does, with 128K tokens against 33K.
How many benchmarks do Llama 4 Maverick and Qwen Max share?
23 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Qwen Max has 23.