Model comparison
Amazon Nova Pro vs DeepSeek-R1-Distill-Qwen-32B
DeepSeek-R1-Distill-Qwen-32B is the stronger model overall, scoring 35.5 to 31.0 on the Noometry Index.
Last verified . 8 shared benchmarks.
DeepSeekSummary
- They share 8 benchmarks with published results for both. Amazon Nova Pro scores higher in 2 categories and DeepSeek-R1-Distill-Qwen-32B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where DeepSeek-R1-Distill-Qwen-32B leads 28.1 to 16.7.
- The biggest single-benchmark swing is LiveBench Math: 38% for Amazon Nova Pro and 59.4% for DeepSeek-R1-Distill-Qwen-32B.
- DeepSeek-R1-Distill-Qwen-32B has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B | |
|---|---|---|
| Provider | Amazon | DeepSeek |
| Noometry Index | 31.0 | 35.5 |
| Released | 2024-12-03 | 2025-01-20 |
| Weights | Proprietary | Open |
| Context window | 300K | — |
| Max output | 10K | — |
| Input $ / M tokens | $0.80 | — |
| Output $ / M tokens | $3.20 | — |
| Results tracked | 38 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1-Distill-Qwen-32B leads
Amazon Nova Pro: 35.1 (#229), DeepSeek-R1-Distill-Qwen-32B: 36.1 (#212)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Coding | 38.1% | 33.7% |
| BigCodeBench Instruct | — | 43.9% |
| LMArena Coding | 1270 | — |
| BigCodeBench Complete | — | 54.9% |
Agentic & Tool Use DeepSeek-R1-Distill-Qwen-32B leads
Amazon Nova Pro: 16.7 (#147), DeepSeek-R1-Distill-Qwen-32B: 28.1 (#94)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
| BALROG | — | 19.5% |
Reasoning Amazon Nova Pro leads
Amazon Nova Pro: 20.0 (#243), DeepSeek-R1-Distill-Qwen-32B: 18.2 (#284)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Reasoning | 32.6% | 52.3% |
| LiveBench Data Analysis | 48.3% | 45.4% |
| Epoch Capabilities Index | 123.8 | 137.44 |
| LiveBench | 43.5% | 45.5% |
| Chess Puzzles | — | 1% |
| LMArena Hard Prompts | 1246 | — |
Math DeepSeek-R1-Distill-Qwen-32B leads
Amazon Nova Pro: 28.5 (#243), DeepSeek-R1-Distill-Qwen-32B: 34.5 (#194)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Math | 38% | 59.4% |
| OTIS Mock AIME 2024-2025 | — | 55.6% |
| Omni-MATH | 24.2% | — |
| LMArena Math | 1252 | — |
Knowledge DeepSeek-R1-Distill-Qwen-32B leads
Amazon Nova Pro: 27.4 (#250), DeepSeek-R1-Distill-Qwen-32B: 35.7 (#182)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| GPQA Diamond | — | 64.1% |
| Humanity's Last Exam | 4.4% | — |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |
| LMArena Expert | 1211 | — |
| MMLU | 82% | — |
Multimodal Not comparable
Amazon Nova Pro: 25.0 (#126), DeepSeek-R1-Distill-Qwen-32B: —
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LMArena Vision | 980 | — |
Multilingual Not comparable
Amazon Nova Pro: 39.7 (#223), DeepSeek-R1-Distill-Qwen-32B: —
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LMArena Non-English | 1234 | — |
| LMArena Chinese | 1244 | — |
| LMArena French | 1271 | — |
| LMArena German | 1243 | — |
| LMArena Japanese | 1200 | — |
| LMArena Korean | 1203 | — |
| LMArena Russian | 1240 | — |
| LMArena Spanish | 1182 | — |
Instruction Following Amazon Nova Pro leads
Amazon Nova Pro: 64.9 (#226), DeepSeek-R1-Distill-Qwen-32B: 61.6 (#243)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Instruction Following | 67.1% | 55.7% |
| IFEval | 81.5% | — |
| LMArena Instruction Following | 1235 | — |
Long Context Not comparable
Amazon Nova Pro: 38.1 (#205), DeepSeek-R1-Distill-Qwen-32B: —
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LMArena Longer Query | 1255 | — |
Writing & Preference DeepSeek-R1-Distill-Qwen-32B leads
Amazon Nova Pro: 43.9 (#226), DeepSeek-R1-Distill-Qwen-32B: 49.6 (#188)
| Benchmark | Amazon Nova Pro | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| LiveBench Language | 37% | 26.8% |
| LMArena Text | 1259 | — |
| LMArena Creative Writing | 1212 | — |
| Short-Story Creative Writing | 60.5% | — |
| WildBench | 77.7% | — |
| LMArena Multi-Turn | 1246 | — |
Frequently asked questions
Is Amazon Nova Pro better than DeepSeek-R1-Distill-Qwen-32B?
DeepSeek-R1-Distill-Qwen-32B is the stronger model overall, scoring 35.5 to 31.0 on the Noometry Index.
Is Amazon Nova Pro or DeepSeek-R1-Distill-Qwen-32B better for coding?
DeepSeek-R1-Distill-Qwen-32B scores higher on coding benchmarks: 36.1 versus 35.1 in the Noometry coding category.
How many benchmarks do Amazon Nova Pro and DeepSeek-R1-Distill-Qwen-32B share?
8 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and DeepSeek-R1-Distill-Qwen-32B has 14.