Model comparison
Deepseek Coder v2 vs Nemotron 3 Super
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 35.9 on the Noometry Index.
Last verified . 15 shared benchmarks.
DeepSeek
NVIDIASummary
- They share 15 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and Nemotron 3 Super in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Nemotron 3 Super leads 55.9 to 38.2.
Side by side
| Deepseek Coder v2 | Nemotron 3 Super | |
|---|---|---|
| Provider | DeepSeek | NVIDIA |
| Noometry Index | 35.9 | 40.1 |
| Released | 2024-06-17 | 2026-03-11 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.08 |
| Output $ / M tokens | — | $0.45 |
| Results tracked | 24 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Super leads
Deepseek Coder v2: 38.1 (#183), Nemotron 3 Super: 39.4 (#158)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Coding | 1251 | 1403 |
| SciCode | — | 36% |
| WeirdML | — | 38% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| ALE-Bench | — | 213.9 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Nemotron 3 Super: 18.7 (#276)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1386 |
| NYT Connections (extended) | — | 15.4% |
| CritPt | — | 3.1% |
| WinoGrande | 83.7% | — |
Math Nemotron 3 Super leads
Deepseek Coder v2: 34.9 (#190), Nemotron 3 Super: 39.6 (#101)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Math | 1241 | 1380 |
| MathArena Final-Answer Competitions | — | 60.4% |
| GSM8K | 94.5% | — |
Knowledge Nemotron 3 Super leads
Deepseek Coder v2: 32.3 (#212), Nemotron 3 Super: 38.9 (#139)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Expert | 1181 | 1398 |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Nemotron 3 Super leads
Deepseek Coder v2: 36.3 (#240), Nemotron 3 Super: 48.4 (#144)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Non-English | 1182 | 1355 |
| LMArena Chinese | 1201 | 1432 |
| LMArena French | 1185 | 1405 |
| LMArena German | 1164 | 1341 |
| LMArena Russian | 1188 | 1336 |
| LMArena Spanish | 1153 | 1417 |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
Instruction Following Nemotron 3 Super leads
Deepseek Coder v2: 61.7 (#242), Nemotron 3 Super: 71.2 (#154)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Instruction Following | 1180 | 1347 |
Long Context Nemotron 3 Super leads
Deepseek Coder v2: 37.0 (#224), Nemotron 3 Super: 41.5 (#139)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Longer Query | 1219 | 1362 |
Writing & Preference Nemotron 3 Super leads
Deepseek Coder v2: 38.2 (#253), Nemotron 3 Super: 55.9 (#140)
| Benchmark | Deepseek Coder v2 | Nemotron 3 Super |
|---|---|---|
| LMArena Text | 1191 | 1378 |
| LMArena Creative Writing | 1120 | 1317 |
| LMArena Multi-Turn | 1177 | 1369 |
Frequently asked questions
Is Deepseek Coder v2 better than Nemotron 3 Super?
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 35.9 on the Noometry Index.
Is Deepseek Coder v2 or Nemotron 3 Super better for coding?
Nemotron 3 Super scores higher on coding benchmarks: 39.4 versus 38.1 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Nemotron 3 Super share?
15 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Nemotron 3 Super has 21.