Model comparison
Llama 4 Scout vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 27.7 on the Noometry Index. Llama 4 Scout costs 5.0× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Llama 4 Scout scores higher in 2 categories and Qwen2.5-Coder-32B in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 33.3 to 19.6.
- The biggest single-benchmark swing is BigCodeBench Complete: 43.1% for Llama 4 Scout and 58% for Qwen2.5-Coder-32B.
- Llama 4 Scout is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Llama 4 Scout accepts more context: 128K tokens versus 33K.
Side by side
| Llama 4 Scout | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 27.7 | 33.4 |
| Released | 2025-04-05 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 128K | 33K |
| Max output | 4K | 29K |
| Input $ / M tokens | $0.10 | $0.66 |
| Output $ / M tokens | $0.30 | $1 |
| Results tracked | 43 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Coder-32B leads
Llama 4 Scout: 20.2 (#339), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| SWE-bench Verified (bash only) | 9.1% | 9% |
| LMArena Coding | 1286 | 1276 |
| BigCodeBench Complete | 43.1% | 58% |
| Aider Polyglot | — | 16.4% |
| SciCode | 17% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Llama 4 Scout: 24.6 (#119), Qwen2.5-Coder-32B: —
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.1% | — |
Reasoning Qwen2.5-Coder-32B leads
Llama 4 Scout: 9.1 (#345), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1266 | 1251 |
| Epoch Capabilities Index | 129.64 | 119.49 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 36.9% | — |
| ARC-AGI-1 | 0.5% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 57.9% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 12% | — |
| ForecastBench | 57.5 | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Llama 4 Scout: 19.6 (#286), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1287 | 1251 |
| OTIS Mock AIME 2024-2025 | 7.8% | — |
| Omni-MATH | 37.3% | — |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 62.3% | — |
| FrontierMath (Feb 2025 set) | 0% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Llama 4 Scout: 31.9 (#217), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1235 | 1221 |
| GPQA Diamond | 51.8% | — |
| MMLU-Pro | 74.2% | — |
| Vectara Hallucination Rate | 7.7% | — |
| GPQA (HELM) | 50.7% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
Llama 4 Scout: 32.2 (#102), Qwen2.5-Coder-32B: —
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1118 | — |
| SpatialViz-Bench | 34.2% | — |
Multilingual Llama 4 Scout leads
Llama 4 Scout: 41.0 (#212), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1252 | 1205 |
| LMArena Chinese | 1255 | 1222 |
| LMArena Russian | 1263 | 1228 |
| LMArena French | 1282 | — |
| LMArena German | 1272 | — |
| LMArena Japanese | 1206 | — |
| LMArena Korean | 1207 | — |
| LMArena Spanish | 1278 | — |
Instruction Following Llama 4 Scout leads
Llama 4 Scout: 65.8 (#217), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1248 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
| IFEval | 81.8% | — |
Long Context Qwen2.5-Coder-32B leads
Llama 4 Scout: 27.5 (#294), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1265 | 1251 |
| Fiction.LiveBench | 36% | — |
Writing & Preference Qwen2.5-Coder-32B leads
Llama 4 Scout: 37.0 (#261), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Llama 4 Scout | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1279 | 1230 |
| LMArena Creative Writing | 1249 | 1174 |
| LMArena Multi-Turn | 1280 | 1222 |
| EQ-Bench Creative Writing | 783 | — |
| WildBench | 78% | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Llama 4 Scout better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 27.7 on the Noometry Index. Llama 4 Scout costs 5.0× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Scout or Qwen2.5-Coder-32B?
Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Llama 4 Scout or Qwen2.5-Coder-32B better for coding?
Qwen2.5-Coder-32B scores higher on coding benchmarks: 22.6 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Llama 4 Scout does, with 128K tokens against 33K.
How many benchmarks do Llama 4 Scout and Qwen2.5-Coder-32B share?
15 benchmarks have published results for both models. Llama 4 Scout has 43 scored results on Noometry and Qwen2.5-Coder-32B has 31.