Model comparison
Llama 4 Maverick vs Qwen2.5 7B Instruct
Llama 4 Maverick is the stronger model overall, scoring 30.6 to 28.6 on the Noometry Index.
Last verified . 12 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Llama 4 Maverick scores higher in 4 categories and Qwen2.5 7B Instruct in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 4 Maverick leads 33.3 to 17.0.
- The biggest single-benchmark swing is GPQA Diamond: 67% for Llama 4 Maverick and 35.5% for Qwen2.5 7B Instruct.
- Both cost about the same: $0.19 input and $0.65 output per million tokens.
- Qwen2.5 7B Instruct accepts more context: 131K tokens versus 128K.
Side by side
| Llama 4 Maverick | Qwen2.5 7B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.6 | 28.6 |
| Released | 2025-04-05 | 2024-09 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.19 | $0.17 |
| Output $ / M tokens | $0.65 | $0.70 |
| Results tracked | 54 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 7B Instruct leads
Llama 4 Maverick: 26.3 (#318), Qwen2.5 7B Instruct: 35.9 (#204)
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| BigCodeBench Instruct | 49.7% | 37.6% |
| BigCodeBench Complete | 61.4% | 46.1% |
| SWE-bench Verified (bash only) | 21% | — |
| Aider Polyglot | 15.6% | — |
| SciCode | 33.1% | — |
| WeirdML | 24.5% | — |
| LMArena Coding | 1302 | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Llama 4 Maverick leads
Llama 4 Maverick: 27.9 (#86), Qwen2.5 7B Instruct: 23.5 (#117)
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
| BALROG | — | 7.8% |
Reasoning Qwen2.5 7B Instruct leads
Llama 4 Maverick: 10.1 (#337), Qwen2.5 7B Instruct: 14.3 (#317)
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| DTBench | 61.9% | 47.7% |
| LMCA | 15.9% | 6.4% |
| Epoch Capabilities Index | 132.2 | 118.44 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 27.7% | — |
| Kagi LLM Benchmark | 55.9% | — |
| NYT Connections (extended) | 8% | — |
| ARC-AGI-1 | 4.4% | — |
| CritPt | 0% | — |
| Chess Puzzles | — | 0% |
| EnigmaEval | 0.6% | — |
| LMArena Hard Prompts | 1281 | — |
| ForecastBench | 57.5 | — |
Math Llama 4 Maverick leads
Llama 4 Maverick: 24.6 (#256), Qwen2.5 7B Instruct: 11.8 (#300)
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 20.6% | 2.5% |
| Omni-MATH | 42.2% | 29.4% |
| LMArena Math | 1299 | — |
| MATH Level 5 | 73% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.3 (#196), Qwen2.5 7B Instruct: 17.0 (#281)
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 67% | 35.5% |
| MMLU-Pro | 81% | 53.9% |
| GPQA (HELM) | 65% | 34.1% |
| Humanity's Last Exam | 5.7% | — |
| Confabulations | 22.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
| LMArena Expert | 1259 | — |
| MMLU | — | 72.9% |
Multimodal Not comparable
Llama 4 Maverick: 30.9 (#100), Qwen2.5 7B Instruct: —
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Vision | 1141 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Not comparable
Llama 4 Maverick: 42.4 (#190), Qwen2.5 7B Instruct: —
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1269 | — |
| LMArena Chinese | 1277 | — |
| LMArena French | 1258 | — |
| LMArena German | 1291 | — |
| LMArena Japanese | 1207 | — |
| LMArena Korean | 1203 | — |
| LMArena Russian | 1286 | — |
| LMArena Spanish | 1293 | — |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.9 (#140), Qwen2.5 7B Instruct: 63.4 (#226)
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | 90.8% | 74.1% |
| LMArena Instruction Following | 1267 | — |
Long Context Not comparable
Llama 4 Maverick: 31.4 (#274), Qwen2.5 7B Instruct: —
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| Fiction.LiveBench | 46.2% | — |
| LMArena Longer Query | 1280 | — |
Writing & Preference Qwen2.5 7B Instruct leads
Llama 4 Maverick: 39.0 (#247), Qwen2.5 7B Instruct: 48.8 (#194)
| Benchmark | Llama 4 Maverick | Qwen2.5 7B Instruct |
|---|---|---|
| WildBench | 80% | 73.1% |
| LMArena Text | 1287 | — |
| LMArena Creative Writing | 1266 | — |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
| LMArena Multi-Turn | 1289 | — |
Frequently asked questions
Is Llama 4 Maverick better than Qwen2.5 7B Instruct?
Llama 4 Maverick is the stronger model overall, scoring 30.6 to 28.6 on the Noometry Index.
Which is cheaper, Llama 4 Maverick or Qwen2.5 7B Instruct?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Qwen2.5 7B Instruct lists at $0.17 and $0.70.
Is Llama 4 Maverick or Qwen2.5 7B Instruct better for coding?
Qwen2.5 7B Instruct scores higher on coding benchmarks: 35.9 versus 26.3 in the Noometry coding category.
Which has the bigger context window?
Qwen2.5 7B Instruct does, with 131K tokens against 128K.
How many benchmarks do Llama 4 Maverick and Qwen2.5 7B Instruct share?
12 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Qwen2.5 7B Instruct has 15.