Model comparison
DeepSeek-V2.5 (Sep 2024) vs Nemotron 3 Super
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 37.6 on the Noometry Index.
Last verified . 15 shared benchmarks.
DeepSeek
NVIDIASummary
- They share 15 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 1 category and Nemotron 3 Super in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Nemotron 3 Super leads 39.4 to 31.7.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super | |
|---|---|---|
| Provider | DeepSeek | NVIDIA |
| Noometry Index | 37.6 | 40.1 |
| Released | 2024-09-06 | 2026-03-11 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.08 |
| Output $ / M tokens | — | $0.45 |
| Results tracked | 22 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Super leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Nemotron 3 Super: 39.4 (#158)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Coding | 1309 | 1403 |
| Aider Polyglot | 17.8% | — |
| SciCode | — | 36% |
| WeirdML | — | 38% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 213.9 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Nemotron 3 Super: 18.7 (#276)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1386 |
| NYT Connections (extended) | — | 15.4% |
| CritPt | — | 3.1% |
Math Nemotron 3 Super leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Nemotron 3 Super: 39.6 (#101)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Math | 1288 | 1380 |
| MathArena Final-Answer Competitions | — | 60.4% |
Knowledge Nemotron 3 Super leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Nemotron 3 Super: 38.9 (#139)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Expert | 1266 | 1398 |
Multilingual Nemotron 3 Super leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Nemotron 3 Super: 48.4 (#144)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Non-English | 1273 | 1355 |
| LMArena Chinese | 1318 | 1432 |
| LMArena French | 1289 | 1405 |
| LMArena German | 1258 | 1341 |
| LMArena Russian | 1289 | 1336 |
| LMArena Spanish | 1248 | 1417 |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
Instruction Following Nemotron 3 Super leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Nemotron 3 Super: 71.2 (#154)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Instruction Following | 1280 | 1347 |
Long Context Nemotron 3 Super leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Nemotron 3 Super: 41.5 (#139)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Longer Query | 1301 | 1362 |
Writing & Preference Nemotron 3 Super leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Nemotron 3 Super: 55.9 (#140)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nemotron 3 Super |
|---|---|---|
| LMArena Text | 1294 | 1378 |
| LMArena Creative Writing | 1285 | 1317 |
| LMArena Multi-Turn | 1297 | 1369 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Nemotron 3 Super?
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Nemotron 3 Super better for coding?
Nemotron 3 Super scores higher on coding benchmarks: 39.4 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Nemotron 3 Super share?
15 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Nemotron 3 Super has 21.