Model comparison
Llama 4 Scout vs Qwen2.5-VL 72B Instruct
Qwen2.5-VL 72B Instruct is the stronger model overall, scoring 29.9 to 27.7 on the Noometry Index. Llama 4 Scout costs 28× less per token, which makes it the better buy when Qwen2.5-VL 72B Instruct's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. Llama 4 Scout scores higher in 1 category and Qwen2.5-VL 72B Instruct in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen2.5-VL 72B Instruct leads 20.7 to 9.1.
- Llama 4 Scout is cheaper at $0.10 / $0.30 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
- Qwen2.5-VL 72B Instruct accepts more context: 131K tokens versus 128K.
Side by side
| Llama 4 Scout | Qwen2.5-VL 72B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 27.7 | 29.9 |
| Released | 2025-04-05 | 2024-09 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.10 | $2.80 |
| Output $ / M tokens | $0.30 | $8.40 |
| Results tracked | 43 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Llama 4 Scout: 20.2 (#339), Qwen2.5-VL 72B Instruct: —
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| SWE-bench Verified (bash only) | 9.1% | — |
| SciCode | 17% | — |
| LMArena Coding | 1286 | — |
| BigCodeBench Complete | 43.1% | — |
Agentic & Tool Use Llama 4 Scout leads
Llama 4 Scout: 24.6 (#119), Qwen2.5-VL 72B Instruct: 18.6 (#144)
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.1% | — |
| OSWorld | — | 5% |
Reasoning Qwen2.5-VL 72B Instruct leads
Llama 4 Scout: 9.1 (#345), Qwen2.5-VL 72B Instruct: 20.7 (#233)
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 36.9% | 36% |
| ARC-AGI-2 | 0% | — |
| ARC-AGI-1 | 0.5% | — |
| CritPt | 0% | — |
| LMArena Hard Prompts | 1266 | — |
| DTBench | 57.9% | — |
| LMCA | 12% | — |
| Epoch Capabilities Index | 129.64 | — |
| ForecastBench | 57.5 | — |
Math Not comparable
Llama 4 Scout: 19.6 (#286), Qwen2.5-VL 72B Instruct: —
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.8% | — |
| Omni-MATH | 37.3% | — |
| LMArena Math | 1287 | — |
| MATH Level 5 | 62.3% | — |
| FrontierMath (Feb 2025 set) | 0% | — |
Knowledge Not comparable
Llama 4 Scout: 31.9 (#217), Qwen2.5-VL 72B Instruct: —
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| GPQA Diamond | 51.8% | — |
| MMLU-Pro | 74.2% | — |
| Vectara Hallucination Rate | 7.7% | — |
| GPQA (HELM) | 50.7% | — |
| LMArena Expert | 1235 | — |
Multimodal Qwen2.5-VL 72B Instruct leads
Llama 4 Scout: 32.2 (#102), Qwen2.5-VL 72B Instruct: 33.5 (#97)
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Vision | 1118 | 1107 |
| SpatialViz-Bench | 34.2% | 33.3% |
| Video-MME | — | 73.5% |
| GeoBench | — | 62% |
Multilingual Not comparable
Llama 4 Scout: 41.0 (#212), Qwen2.5-VL 72B Instruct: —
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Non-English | 1252 | — |
| LMArena Chinese | 1255 | — |
| LMArena French | 1282 | — |
| LMArena German | 1272 | — |
| LMArena Japanese | 1206 | — |
| LMArena Korean | 1207 | — |
| LMArena Russian | 1263 | — |
| LMArena Spanish | 1278 | — |
Instruction Following Not comparable
Llama 4 Scout: 65.8 (#217), Qwen2.5-VL 72B Instruct: —
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| IFEval | 81.8% | — |
| LMArena Instruction Following | 1248 | — |
Long Context Not comparable
Llama 4 Scout: 27.5 (#294), Qwen2.5-VL 72B Instruct: —
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Fiction.LiveBench | 36% | — |
| LMArena Longer Query | 1265 | — |
Writing & Preference Not comparable
Llama 4 Scout: 37.0 (#261), Qwen2.5-VL 72B Instruct: —
| Benchmark | Llama 4 Scout | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Text | 1279 | — |
| LMArena Creative Writing | 1249 | — |
| EQ-Bench Creative Writing | 783 | — |
| WildBench | 78% | — |
| LMArena Multi-Turn | 1280 | — |
Frequently asked questions
Is Llama 4 Scout better than Qwen2.5-VL 72B Instruct?
Qwen2.5-VL 72B Instruct is the stronger model overall, scoring 29.9 to 27.7 on the Noometry Index. Llama 4 Scout costs 28× less per token, which makes it the better buy when Qwen2.5-VL 72B Instruct's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Scout or Qwen2.5-VL 72B Instruct?
Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.
Which has the bigger context window?
Qwen2.5-VL 72B Instruct does, with 131K tokens against 128K.
How many benchmarks do Llama 4 Scout and Qwen2.5-VL 72B Instruct share?
3 benchmarks have published results for both models. Llama 4 Scout has 43 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.