Model comparison
Nemotron 3 Super vs Qwen2.5 72B Instruct
Nemotron 3 Super is the stronger model overall, scoring 39.6 to 31.5 on the Noometry Index.
Last verified . 16 shared benchmarks.
NVIDIA
Alibaba (Qwen)Summary
- They share 16 benchmarks with published results for both. Nemotron 3 Super scores higher in 7 categories and Qwen2.5 72B Instruct in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Nemotron 3 Super leads 37.6 to 18.1.
- The biggest single-benchmark swing is WeirdML: 38% for Nemotron 3 Super and 16% for Qwen2.5 72B Instruct.
- Nemotron 3 Super is cheaper at $0.08 / $0.45 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Nemotron 3 Super accepts more context: 262K tokens versus 131K.
Side by side
| Nemotron 3 Super | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | NVIDIA | Alibaba (Qwen) |
| Noometry Index | 39.6 | 31.5 |
| Released | 2026-03-11 | 2024-09 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 131K | 8K |
| Input $ / M tokens | $0.08 | $1.40 |
| Output $ / M tokens | $0.45 | $5.60 |
| Results tracked | 21 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Super leads
Nemotron 3 Super: 38.9 (#153), Qwen2.5 72B Instruct: 32.8 (#256)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 38% | 16% |
| LMArena Coding | 1403 | 1293 |
| SciCode | 36% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| ALE-Bench | 213.9 | — |
Agentic & Tool Use Not comparable
Nemotron 3 Super: —, Qwen2.5 72B Instruct: 22.0 (#124)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Nemotron 3 Super: 17.9 (#274), Qwen2.5 72B Instruct: 21.4 (#197)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1386 | 1271 |
| NYT Connections (extended) | 15.4% | — |
| CritPt | 3.1% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129.01 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Nemotron 3 Super leads
Nemotron 3 Super: 37.6 (#95), Qwen2.5 72B Instruct: 18.1 (#281)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Math | 1381 | 1283 |
| MathArena Final-Answer Competitions | 60.4% | — |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |
Knowledge Nemotron 3 Super leads
Nemotron 3 Super: 38.2 (#136), Qwen2.5 72B Instruct: 26.9 (#247)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Expert | 1397 | 1245 |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multilingual Nemotron 3 Super leads
Nemotron 3 Super: 48.6 (#139), Qwen2.5 72B Instruct: 41.2 (#208)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1356 | 1252 |
| LMArena Chinese | 1430 | 1272 |
| LMArena French | 1404 | 1279 |
| LMArena German | 1341 | 1234 |
| LMArena Russian | 1336 | 1264 |
| LMArena Spanish | 1417 | 1255 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
Instruction Following Nemotron 3 Super leads
Nemotron 3 Super: 71.4 (#149), Qwen2.5 72B Instruct: 65.7 (#216)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1347 | 1254 |
| IFEval | — | 80.6% |
Long Context Nemotron 3 Super leads
Nemotron 3 Super: 41.5 (#134), Qwen2.5 72B Instruct: 38.9 (#183)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1362 | 1282 |
Writing & Preference Nemotron 3 Super leads
Nemotron 3 Super: 56.1 (#135), Qwen2.5 72B Instruct: 46.9 (#210)
| Benchmark | Nemotron 3 Super | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1378 | 1269 |
| LMArena Creative Writing | 1317 | 1221 |
| LMArena Multi-Turn | 1369 | 1272 |
| WildBench | — | 80.2% |
Frequently asked questions
Is Nemotron 3 Super better than Qwen2.5 72B Instruct?
Nemotron 3 Super is the stronger model overall, scoring 39.6 to 31.5 on the Noometry Index.
Which is cheaper, Nemotron 3 Super or Qwen2.5 72B Instruct?
Nemotron 3 Super is cheaper. It lists at $0.08 per million input tokens and $0.45 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is Nemotron 3 Super or Qwen2.5 72B Instruct better for coding?
Nemotron 3 Super scores higher on coding benchmarks: 38.9 versus 32.8 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Super does, with 262K tokens against 131K.
How many benchmarks do Nemotron 3 Super and Qwen2.5 72B Instruct share?
16 benchmarks have published results for both models. Nemotron 3 Super has 21 scored results on Noometry and Qwen2.5 72B Instruct has 43.