Model comparison
Amazon Nova Pro vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 31.0 on the Noometry Index.
Last verified . 21 shared benchmarks.
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. Amazon Nova Pro scores higher in 5 categories and Qwen2.5-Coder-32B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Amazon Nova Pro leads 35.1 to 22.6.
- The biggest single-benchmark swing is LiveBench Coding: 38.1% for Amazon Nova Pro and 56.9% for Qwen2.5-Coder-32B.
- Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $0.80 / $3.20 for Amazon Nova Pro.
- Amazon Nova Pro accepts more context: 300K tokens versus 33K.
- Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Amazon | Alibaba (Qwen) |
| Noometry Index | 31.0 | 33.4 |
| Released | 2024-12-03 | 2024-09-18 |
| Weights | Proprietary | Open |
| Context window | 300K | 33K |
| Max output | 10K | 29K |
| Input $ / M tokens | $0.80 | $0.66 |
| Output $ / M tokens | $3.20 | $1 |
| Results tracked | 38 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Amazon Nova Pro leads
Amazon Nova Pro: 35.1 (#229), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Coding | 38.1% | 56.9% |
| LMArena Coding | 1270 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Amazon Nova Pro: 16.7 (#147), Qwen2.5-Coder-32B: —
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
Reasoning Qwen2.5-Coder-32B leads
Amazon Nova Pro: 20.0 (#243), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Reasoning | 32.6% | 42.1% |
| LMArena Hard Prompts | 1246 | 1251 |
| LiveBench Data Analysis | 48.3% | 49.9% |
| Epoch Capabilities Index | 123.8 | 119.49 |
| LiveBench | 43.5% | 46.2% |
| HellaSwag | — | 83% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Amazon Nova Pro: 28.5 (#243), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Math | 38% | 46.6% |
| LMArena Math | 1252 | 1251 |
| Omni-MATH | 24.2% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Amazon Nova Pro: 27.4 (#250), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1211 | 1221 |
| MMLU | 82% | 79.1% |
| Humanity's Last Exam | 4.4% | — |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |
| ARC (AI2) Challenge | — | 70.5% |
Multimodal Not comparable
Amazon Nova Pro: 25.0 (#126), Qwen2.5-Coder-32B: —
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 980 | — |
Multilingual Amazon Nova Pro leads
Amazon Nova Pro: 39.7 (#223), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1234 | 1205 |
| LMArena Chinese | 1244 | 1222 |
| LMArena Russian | 1240 | 1228 |
| LMArena French | 1271 | — |
| LMArena German | 1243 | — |
| LMArena Japanese | 1200 | — |
| LMArena Korean | 1203 | — |
| LMArena Spanish | 1182 | — |
Instruction Following Amazon Nova Pro leads
Amazon Nova Pro: 64.9 (#226), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Instruction Following | 67.1% | 58.7% |
| LMArena Instruction Following | 1235 | 1223 |
| IFEval | 81.5% | — |
Long Context Too close to call
Amazon Nova Pro: 38.1 (#205), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1255 | 1251 |
Writing & Preference Amazon Nova Pro leads
Amazon Nova Pro: 43.9 (#226), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Amazon Nova Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1259 | 1230 |
| LMArena Creative Writing | 1212 | 1174 |
| LMArena Multi-Turn | 1246 | 1222 |
| LiveBench Language | 37% | 23.3% |
| Short-Story Creative Writing | 60.5% | — |
| WildBench | 77.7% | — |
Frequently asked questions
Is Amazon Nova Pro better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 31.0 on the Noometry Index.
Which is cheaper, Amazon Nova Pro or Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; Amazon Nova Pro lists at $0.80 and $3.20.
Is Amazon Nova Pro or Qwen2.5-Coder-32B better for coding?
Amazon Nova Pro scores higher on coding benchmarks: 35.1 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Pro does, with 300K tokens against 33K.
How many benchmarks do Amazon Nova Pro and Qwen2.5-Coder-32B share?
21 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Qwen2.5-Coder-32B has 31.