Model comparison
Amazon Nova Pro vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 31.0 on the Noometry Index.
Last verified . 27 shared benchmarks.
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. Amazon Nova Pro scores higher in 1 category and Qwen3 235B-A22B in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 235B-A22B leads 49.6 to 27.4.
- The biggest single-benchmark swing is Omni-MATH: 24.2% for Amazon Nova Pro and 71.8% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $0.80 / $3.20 for Amazon Nova Pro.
- Amazon Nova Pro accepts more context: 300K tokens versus 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Amazon | Alibaba (Qwen) |
| Noometry Index | 31.0 | 43.5 |
| Released | 2024-12-03 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 300K | 131K |
| Max output | 10K | 16K |
| Input $ / M tokens | $0.80 | $0.70 |
| Output $ / M tokens | $3.20 | $2.80 |
| Results tracked | 38 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Amazon Nova Pro: 35.1 (#229), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Coding | 1270 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| WeirdML | — | 41% |
| LiveBench Coding | 38.1% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Amazon Nova Pro: 16.7 (#147), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | 52.1% |
| TheAgentCompany | 1.7% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning Amazon Nova Pro leads
Amazon Nova Pro: 20.0 (#243), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1433 |
| Epoch Capabilities Index | 123.8 | 143.85 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Chess Puzzles | — | 12% |
| LiveBench Reasoning | 32.6% | — |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 29.3% |
| ForecastBench | — | 59.7 |
| LiveBench | 43.5% | — |
Math Qwen3 235B-A22B leads
Amazon Nova Pro: 28.5 (#243), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| Omni-MATH | 24.2% | 71.8% |
| LMArena Math | 1252 | 1432 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| LiveBench Math | 38% | — |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Amazon Nova Pro: 27.4 (#250), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| MMLU-Pro | 67.3% | 84.4% |
| Confabulations | 30.1% | 15.6% |
| Vectara Hallucination Rate | 5.1% | 9.3% |
| GPQA (HELM) | 44.6% | 72.7% |
| LMArena Expert | 1211 | 1463 |
| GPQA Diamond | — | 80.1% |
| Humanity's Last Exam | 4.4% | — |
| SimpleQA Verified | — | 40.4% |
| MMLU | 82% | — |
Multimodal Not comparable
Amazon Nova Pro: 25.0 (#126), Qwen3 235B-A22B: —
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 980 | — |
Multilingual Qwen3 235B-A22B leads
Amazon Nova Pro: 39.7 (#223), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1234 | 1409 |
| LMArena Chinese | 1244 | 1481 |
| LMArena French | 1271 | 1445 |
| LMArena German | 1243 | 1433 |
| LMArena Japanese | 1200 | 1399 |
| LMArena Korean | 1203 | 1391 |
| LMArena Russian | 1240 | 1411 |
| LMArena Spanish | 1182 | 1430 |
Instruction Following Qwen3 235B-A22B leads
Amazon Nova Pro: 64.9 (#226), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| IFEval | 81.5% | 83.5% |
| LMArena Instruction Following | 1235 | 1408 |
| LiveBench Instruction Following | 67.1% | — |
Long Context Qwen3 235B-A22B leads
Amazon Nova Pro: 38.1 (#205), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1255 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Qwen3 235B-A22B leads
Amazon Nova Pro: 43.9 (#226), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Amazon Nova Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1259 | 1419 |
| LMArena Creative Writing | 1212 | 1384 |
| Short-Story Creative Writing | 60.5% | 83% |
| WildBench | 77.7% | 86.6% |
| LMArena Multi-Turn | 1246 | 1432 |
| EQ-Bench Creative Writing | — | 1366 |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 31.0 on the Noometry Index.
Which is cheaper, Amazon Nova Pro or Qwen3 235B-A22B?
Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Amazon Nova Pro lists at $0.80 and $3.20.
Is Amazon Nova Pro or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 35.1 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Pro does, with 300K tokens against 131K.
How many benchmarks do Amazon Nova Pro and Qwen3 235B-A22B share?
27 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Qwen3 235B-A22B has 49.