Model comparison
DeepSeek-V3 vs Nemotron 3 Super
DeepSeek-V3 and Nemotron 3 Super score almost the same on the Noometry Index (39.5 vs 40.1), so choose on price, context window or the category you care about most.
Last verified . 18 shared benchmarks.
DeepSeek
NVIDIASummary
- They share 18 benchmarks with published results for both. DeepSeek-V3 scores higher in 5 categories and Nemotron 3 Super in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Nemotron 3 Super leads 39.6 to 32.1.
- Nemotron 3 Super is cheaper at $0.08 / $0.45 per million input/output tokens, against $0.24 / $0.90 for DeepSeek-V3.
- Nemotron 3 Super accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3 | Nemotron 3 Super | |
|---|---|---|
| Provider | DeepSeek | NVIDIA |
| Noometry Index | 39.5 | 40.1 |
| Released | 2024-12-26 | 2026-03-11 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 164K | 131K |
| Input $ / M tokens | $0.24 | $0.08 |
| Output $ / M tokens | $0.90 | $0.45 |
| Results tracked | 60 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3 leads
DeepSeek-V3: 42.3 (#106), Nemotron 3 Super: 39.4 (#158)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| SciCode | 35.8% | 36% |
| WeirdML | 36.1% | 38% |
| LMArena Coding | 1368 | 1403 |
| Aider Polyglot | 55.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| ALE-Bench | — | 213.9 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, Nemotron 3 Super: —
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| METR Time Horizons | 49.6% | — |
Reasoning DeepSeek-V3 leads
DeepSeek-V3: 20.5 (#236), Nemotron 3 Super: 18.7 (#276)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| CritPt | 0% | 3.1% |
| LMArena Hard Prompts | 1365 | 1386 |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| NYT Connections (extended) | — | 15.4% |
| LiveBench Reasoning | 65.8% | — |
| DTBench | 64.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 15.5% | — |
| BIG-Bench Hard | 87.5% | — |
| Epoch Capabilities Index | 135.94 | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |
Math Nemotron 3 Super leads
DeepSeek-V3: 32.1 (#219), Nemotron 3 Super: 39.6 (#101)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| LMArena Math | 1373 | 1380 |
| MathArena Final-Answer Competitions | — | 60.4% |
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge Nemotron 3 Super leads
DeepSeek-V3: 37.5 (#155), Nemotron 3 Super: 38.9 (#139)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| LMArena Expert | 1351 | 1398 |
| GPQA Diamond | 67.6% | — |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 53.8% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 87.2% | — |
| TriviaQA | 82.9% | — |
Multilingual Too close to call
DeepSeek-V3: 48.5 (#143), Nemotron 3 Super: 48.4 (#144)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| LMArena Non-English | 1358 | 1355 |
| LMArena Chinese | 1391 | 1432 |
| LMArena French | 1385 | 1405 |
| LMArena German | 1374 | 1341 |
| LMArena Russian | 1373 | 1336 |
| LMArena Spanish | 1358 | 1417 |
| LMArena Japanese | 1333 | — |
| LMArena Korean | 1319 | — |
Instruction Following DeepSeek-V3 leads
DeepSeek-V3: 72.8 (#130), Nemotron 3 Super: 71.2 (#154)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| LMArena Instruction Following | 1345 | 1347 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
Long Context Nemotron 3 Super leads
DeepSeek-V3: 34.0 (#253), Nemotron 3 Super: 41.5 (#139)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| LMArena Longer Query | 1352 | 1362 |
| Fiction.LiveBench | 50% | — |
Writing & Preference DeepSeek-V3 leads
DeepSeek-V3: 57.4 (#130), Nemotron 3 Super: 55.9 (#140)
| Benchmark | DeepSeek-V3 | Nemotron 3 Super |
|---|---|---|
| LMArena Text | 1375 | 1378 |
| LMArena Creative Writing | 1364 | 1317 |
| LMArena Multi-Turn | 1389 | 1369 |
| Short-Story Creative Writing | 77% | — |
| EQ-Bench Creative Writing | 1472 | — |
| WildBench | 83% | — |
| LiveBench Language | 49.1% | — |
Frequently asked questions
Is DeepSeek-V3 better than Nemotron 3 Super?
DeepSeek-V3 and Nemotron 3 Super score almost the same on the Noometry Index (39.5 vs 40.1), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3 or Nemotron 3 Super?
Nemotron 3 Super is cheaper. It lists at $0.08 per million input tokens and $0.45 per million output tokens; DeepSeek-V3 lists at $0.24 and $0.90.
Is DeepSeek-V3 or Nemotron 3 Super better for coding?
DeepSeek-V3 scores higher on coding benchmarks: 42.3 versus 39.4 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Super does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3 and Nemotron 3 Super share?
18 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Nemotron 3 Super has 21.