Model comparison
Nemotron 3 Ultra vs Qwen2-72B
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 30.0 on the Noometry Index.
Last verified . 19 shared benchmarks.
NVIDIA
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Nemotron 3 Ultra scores higher in 9 categories and Qwen2-72B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Nemotron 3 Ultra leads 52.5 to 21.2.
- The biggest single-benchmark swing is GPQA Diamond: 85.4% for Nemotron 3 Ultra and 40.8% for Qwen2-72B.
Side by side
| Nemotron 3 Ultra | Qwen2-72B | |
|---|---|---|
| Provider | NVIDIA | Alibaba (Qwen) |
| Noometry Index | 42.5 | 30.0 |
| Released | 2026-06-04 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.60 | — |
| Output $ / M tokens | $2.40 | — |
| Results tracked | 30 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Ultra leads
Nemotron 3 Ultra: 38.1 (#182), Qwen2-72B: 29.1 (#310)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| WeirdML | 43.5% | 11.3% |
| LMArena Coding | 1468 | 1196 |
| FrontierCode | 13.6% | — |
| SciCode | 40.3% | — |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Nemotron 3 Ultra leads
Nemotron 3 Ultra: 23.4 (#126), Qwen2-72B: 17.0 (#146)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| APEX-Agents | 22.7% | — |
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Nemotron 3 Ultra leads
Nemotron 3 Ultra: 31.0 (#82), Qwen2-72B: 23.2 (#181)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1452 | 1191 |
| Epoch Capabilities Index | 146.17 | 125.28 |
| CritPt | 3.1% | — |
| Chess Puzzles | 12% | — |
| Mystery Game Puzzles | 20% | — |
| DTBench | 90.1% | — |
| LMCA | 36.9% | — |
Math Nemotron 3 Ultra leads
Nemotron 3 Ultra: 35.3 (#187), Qwen2-72B: 30.2 (#236)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| LMArena Math | 1457 | 1235 |
| OTIS Mock AIME 2024-2025 | 86.7% | — |
| ProofBench | 2% | — |
| MATH Level 5 | — | 39.1% |
Knowledge Nemotron 3 Ultra leads
Nemotron 3 Ultra: 52.5 (#61), Qwen2-72B: 21.2 (#275)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 85.4% | 40.8% |
| LMArena Expert | 1472 | 1171 |
| MMLU | — | 82.4% |
Multilingual Nemotron 3 Ultra leads
Nemotron 3 Ultra: 53.5 (#64), Qwen2-72B: 35.9 (#244)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1427 | 1176 |
| LMArena Chinese | 1497 | 1240 |
| LMArena French | 1472 | 1170 |
| LMArena German | 1471 | 1151 |
| LMArena Korean | 1386 | 1083 |
| LMArena Russian | 1417 | 1169 |
| LMArena Spanish | 1454 | 1169 |
| LMArena Japanese | — | 1111 |
Instruction Following Nemotron 3 Ultra leads
Nemotron 3 Ultra: 74.7 (#88), Qwen2-72B: 61.7 (#241)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1418 | 1181 |
Long Context Nemotron 3 Ultra leads
Nemotron 3 Ultra: 43.9 (#81), Qwen2-72B: 36.1 (#235)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1437 | 1192 |
Writing & Preference Nemotron 3 Ultra leads
Nemotron 3 Ultra: 66.3 (#36), Qwen2-72B: 40.8 (#241)
| Benchmark | Nemotron 3 Ultra | Qwen2-72B |
|---|---|---|
| LMArena Text | 1445 | 1203 |
| LMArena Creative Writing | 1398 | 1181 |
| LMArena Multi-Turn | 1411 | 1196 |
| EQ-Bench Creative Writing | 1692 | — |
Frequently asked questions
Is Nemotron 3 Ultra better than Qwen2-72B?
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 30.0 on the Noometry Index.
Is Nemotron 3 Ultra or Qwen2-72B better for coding?
Nemotron 3 Ultra scores higher on coding benchmarks: 38.1 versus 29.1 in the Noometry coding category.
How many benchmarks do Nemotron 3 Ultra and Qwen2-72B share?
19 benchmarks have published results for both models. Nemotron 3 Ultra has 30 scored results on Noometry and Qwen2-72B has 26.