Model comparison
DeepSeek-R1 vs Nemotron 3 Ultra
DeepSeek-R1 and Nemotron 3 Ultra score almost the same on the Noometry Index (42.3 vs 42.5), so choose on price, context window or the category you care about most.
Last verified . 23 shared benchmarks.
DeepSeek
NVIDIASummary
- They share 23 benchmarks with published results for both. DeepSeek-R1 scores higher in 4 categories and Nemotron 3 Ultra in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Nemotron 3 Ultra leads 31.0 to 18.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.4% for DeepSeek-R1 and 86.7% for Nemotron 3 Ultra.
- Both cost about the same: $0.50 input and $2.15 output per million tokens.
- Nemotron 3 Ultra accepts more context: 262K tokens versus 164K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1 | Nemotron 3 Ultra | |
|---|---|---|
| Provider | DeepSeek | NVIDIA |
| Noometry Index | 42.3 | 42.5 |
| Released | 2025-01-20 | 2026-06-04 |
| Weights | Proprietary | Open |
| Context window | 164K | 262K |
| Max output | 64K | 128K |
| Input $ / M tokens | $0.50 | $0.50 |
| Output $ / M tokens | $2.15 | $2.20 |
| Results tracked | 52 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| SciCode | 35.7% | 40.3% |
| WeirdML | 41.6% | 43.5% |
| LMArena Coding | 1427 | 1468 |
| FrontierCode | — | 13.6% |
| Aider Polyglot | 71.4% | — |
| LiveBench Coding | 66.7% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use DeepSeek-R1 leads
DeepSeek-R1: 30.7 (#75), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | — | 22.7% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
Reasoning Nemotron 3 Ultra leads
DeepSeek-R1: 18.6 (#278), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 1.1% | 3.1% |
| LMArena Hard Prompts | 1416 | 1452 |
| Epoch Capabilities Index | 141.29 | 146.17 |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 40.8% | — |
| Kagi LLM Benchmark | 69.4% | — |
| ARC-AGI-1 | 21.2% | — |
| Chess Puzzles | — | 12% |
| LiveBench Reasoning | 83.2% | — |
| Mystery Game Puzzles | — | 20% |
| DTBench | — | 90.1% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 36.9% |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 86.7% |
| LMArena Math | 1400 | 1457 |
| ProofBench | — | 2% |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
Knowledge Nemotron 3 Ultra leads
DeepSeek-R1: 44.5 (#87), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 76.3% | 85.4% |
| LMArena Expert | 1394 | 1472 |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
Multilingual Nemotron 3 Ultra leads
DeepSeek-R1: 52.4 (#85), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1412 | 1427 |
| LMArena Chinese | 1442 | 1497 |
| LMArena French | 1417 | 1472 |
| LMArena German | 1404 | 1471 |
| LMArena Korean | 1360 | 1386 |
| LMArena Russian | 1423 | 1417 |
| LMArena Spanish | 1411 | 1454 |
| LMArena Japanese | 1391 | — |
Instruction Following Nemotron 3 Ultra leads
DeepSeek-R1: 72.0 (#143), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1382 | 1418 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1391 | 1437 |
| Fiction.LiveBench | 75% | — |
Writing & Preference Nemotron 3 Ultra leads
DeepSeek-R1: 61.4 (#88), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | DeepSeek-R1 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1428 | 1445 |
| LMArena Creative Writing | 1405 | 1398 |
| EQ-Bench Creative Writing | 1500 | 1692 |
| LMArena Multi-Turn | 1405 | 1411 |
| Short-Story Creative Writing | 83% | — |
| WildBench | 82.8% | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than Nemotron 3 Ultra?
DeepSeek-R1 and Nemotron 3 Ultra score almost the same on the Noometry Index (42.3 vs 42.5), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-R1 or Nemotron 3 Ultra?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Nemotron 3 Ultra lists at $0.50 and $2.20.
Is DeepSeek-R1 or Nemotron 3 Ultra better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Ultra does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-R1 and Nemotron 3 Ultra share?
23 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Nemotron 3 Ultra has 30.