Model comparison
Amazon Nova Pro vs Qwen3.5 27B
Qwen3.5 27B is the stronger model overall, scoring 41.4 to 30.5 on the Noometry Index.
Last verified . 19 shared benchmarks.
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and Qwen3.5 27B in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.5 27B leads 59.5 to 44.1.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 5.1% for Amazon Nova Pro and 12.1% for Qwen3.5 27B.
- Qwen3.5 27B is cheaper at $0.30 / $2.40 per million input/output tokens, against $0.80 / $3.20 for Amazon Nova Pro.
- Amazon Nova Pro accepts more context: 300K tokens versus 262K.
- Qwen3.5 27B has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | Qwen3.5 27B | |
|---|---|---|
| Provider | Amazon | Alibaba (Qwen) |
| Noometry Index | 30.5 | 41.4 |
| Released | 2024-12-03 | 2026-02-23 |
| Weights | Proprietary | Open |
| Context window | 300K | 262K |
| Max output | 10K | 66K |
| Input $ / M tokens | $0.80 | $0.30 |
| Output $ / M tokens | $3.20 | $2.40 |
| Results tracked | 38 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 27B leads
Amazon Nova Pro: 34.6 (#224), Qwen3.5 27B: 38.6 (#159)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Coding | 1270 | 1427 |
| LMArena WebDev | — | 1357 |
| WeirdML | — | 39.5% |
| LiveBench Coding | 38.1% | — |
| ALE-Bench | — | 349.45 |
Agentic & Tool Use Not comparable
Amazon Nova Pro: 16.6 (#139), Qwen3.5 27B: —
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
| Vending-Bench 2 | — | 201.98 |
Reasoning Qwen3.5 27B leads
Amazon Nova Pro: 19.2 (#240), Qwen3.5 27B: 26.6 (#116)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1414 |
| NYT Connections (extended) | — | 47.9% |
| Thematic Generalization | — | 45.5% |
| LiveBench Reasoning | 32.6% | — |
| DTBench | — | 82.4% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 34% |
| Epoch Capabilities Index | 123.75 | — |
| LiveBench | 43.5% | — |
Math Qwen3.5 27B leads
Amazon Nova Pro: 26.4 (#240), Qwen3.5 27B: 36.8 (#103)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1252 | 1428 |
| MathArena Final-Answer Competitions | — | 56.7% |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |
Knowledge Qwen3.5 27B leads
Amazon Nova Pro: 27.1 (#246), Qwen3.5 27B: 37.4 (#149)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| Vectara Hallucination Rate | 5.1% | 12.1% |
| LMArena Expert | 1211 | 1428 |
| Humanity's Last Exam | 4.4% | — |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| GPQA (HELM) | 44.6% | — |
| MMLU | 82% | — |
Multimodal Qwen3.5 27B leads
Amazon Nova Pro: 24.5 (#121), Qwen3.5 27B: 38.7 (#56)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | 980 | 1240 |
Multilingual Qwen3.5 27B leads
Amazon Nova Pro: 39.9 (#218), Qwen3.5 27B: 51.0 (#112)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1234 | 1390 |
| LMArena Chinese | 1244 | 1478 |
| LMArena French | 1270 | 1409 |
| LMArena German | 1243 | 1393 |
| LMArena Japanese | 1200 | 1345 |
| LMArena Korean | 1203 | 1359 |
| LMArena Russian | 1240 | 1389 |
| LMArena Spanish | 1182 | 1408 |
Instruction Following Qwen3.5 27B leads
Amazon Nova Pro: 65.1 (#221), Qwen3.5 27B: 73.7 (#114)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1235 | 1393 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |
Long Context Qwen3.5 27B leads
Amazon Nova Pro: 38.1 (#199), Qwen3.5 27B: 43.1 (#103)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1255 | 1413 |
Writing & Preference Qwen3.5 27B leads
Amazon Nova Pro: 44.1 (#221), Qwen3.5 27B: 59.5 (#106)
| Benchmark | Amazon Nova Pro | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1259 | 1409 |
| LMArena Creative Writing | 1212 | 1362 |
| LMArena Multi-Turn | 1245 | 1410 |
| Short-Story Creative Writing | 60.5% | — |
| WildBench | 77.7% | — |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than Qwen3.5 27B?
Qwen3.5 27B is the stronger model overall, scoring 41.4 to 30.5 on the Noometry Index.
Which is cheaper, Amazon Nova Pro or Qwen3.5 27B?
Qwen3.5 27B is cheaper. It lists at $0.30 per million input tokens and $2.40 per million output tokens; Amazon Nova Pro lists at $0.80 and $3.20.
Is Amazon Nova Pro or Qwen3.5 27B better for coding?
Qwen3.5 27B scores higher on coding benchmarks: 38.6 versus 34.6 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Pro does, with 300K tokens against 262K.
How many benchmarks do Amazon Nova Pro and Qwen3.5 27B share?
19 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Qwen3.5 27B has 28.