Model comparison
Amazon Nova Micro vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 30.4 on the Noometry Index. Amazon Nova Micro costs 40× less per token, which makes it the better buy when Qwen2.5 72B Instruct's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Amazon Nova Micro scores higher in 2 categories and Qwen2.5 72B Instruct in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Qwen2.5 72B Instruct leads 65.5 to 56.3.
- The biggest single-benchmark swing is MMLU-Pro: 51.1% for Amazon Nova Micro and 63.1% for Qwen2.5 72B Instruct.
- Amazon Nova Micro is cheaper at $0.035 / $0.14 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct accepts more context: 131K tokens versus 128K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Micro | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Amazon | Alibaba (Qwen) |
| Noometry Index | 30.4 | 31.9 |
| Released | 2024-12-03 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 128K | 131K |
| Max output | 10K | 8K |
| Input $ / M tokens | $0.035 | $1.40 |
| Output $ / M tokens | $0.14 | $5.60 |
| Results tracked | 32 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
Amazon Nova Micro: 30.5 (#295), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1218 | 1292 |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| LiveBench Coding | 20.2% | — |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Too close to call
Amazon Nova Micro: 22.1 (#132), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 22.3% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Amazon Nova Micro: 17.4 (#294), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1271 |
| LiveBench Reasoning | 25.1% | — |
| DTBench | — | 62.9% |
| LiveBench Data Analysis | 34% | — |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| LiveBench | 29.6% | — |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Amazon Nova Micro leads
Amazon Nova Micro: 26.9 (#254), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| Omni-MATH | 21.4% | 33% |
| LMArena Math | 1206 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| LiveBench Math | 34.5% | — |
| MATH Level 5 | — | 63.2% |
Knowledge Amazon Nova Micro leads
Amazon Nova Micro: 29.6 (#237), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| MMLU-Pro | 51.1% | 63.1% |
| GPQA (HELM) | 38.3% | 42.6% |
| LMArena Expert | 1184 | 1245 |
| MMLU | 70.8% | 85.3% |
| GPQA Diamond | — | 49.1% |
| Confabulations | — | 19.1% |
| Vectara Hallucination Rate | 5.5% | — |
| ARC (AI2) Challenge | — | 94.5% |
| TriviaQA | — | 71.9% |
Multilingual Qwen2.5 72B Instruct leads
Amazon Nova Micro: 36.5 (#239), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1186 | 1252 |
| LMArena Chinese | 1209 | 1272 |
| LMArena French | 1238 | 1280 |
| LMArena German | 1192 | 1234 |
| LMArena Japanese | 1154 | 1180 |
| LMArena Korean | 1150 | 1188 |
| LMArena Russian | 1185 | 1264 |
| LMArena Spanish | 1225 | 1256 |
Instruction Following Qwen2.5 72B Instruct leads
Amazon Nova Micro: 56.3 (#272), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 76% | 80.6% |
| LMArena Instruction Following | 1174 | 1254 |
| LiveBench Instruction Following | 48% | — |
Long Context Qwen2.5 72B Instruct leads
Amazon Nova Micro: 36.5 (#229), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1205 | 1282 |
Writing & Preference Qwen2.5 72B Instruct leads
Amazon Nova Micro: 39.5 (#247), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Amazon Nova Micro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1208 | 1269 |
| LMArena Creative Writing | 1172 | 1221 |
| WildBench | 74.3% | 80.2% |
| LMArena Multi-Turn | 1178 | 1272 |
| LiveBench Language | 15.8% | — |
Frequently asked questions
Is Amazon Nova Micro better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 30.4 on the Noometry Index. Amazon Nova Micro costs 40× less per token, which makes it the better buy when Qwen2.5 72B Instruct's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Micro or Qwen2.5 72B Instruct?
Amazon Nova Micro is cheaper. It lists at $0.035 per million input tokens and $0.14 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is Amazon Nova Micro or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 33.2 versus 30.5 in the Noometry coding category.
Which has the bigger context window?
Qwen2.5 72B Instruct does, with 131K tokens against 128K.
How many benchmarks do Amazon Nova Micro and Qwen2.5 72B Instruct share?
23 benchmarks have published results for both models. Amazon Nova Micro has 32 scored results on Noometry and Qwen2.5 72B Instruct has 43.