Model comparison
Nemotron 3 Super vs Sonar
Nemotron 3 Super is the stronger model overall, scoring 39.6 to 37.8 on the Noometry Index.
Last verified . 0 shared benchmarks.
NVIDIA
PerplexitySummary
- The widest gap is in math, where Nemotron 3 Super leads 37.6 to 31.3.
- Nemotron 3 Super is cheaper at $0.08 / $0.45 per million input/output tokens, against $1 / $1 for Sonar.
- Nemotron 3 Super accepts more context: 262K tokens versus 128K.
- Nemotron 3 Super has downloadable open weights; the other is API-only.
Side by side
| Nemotron 3 Super | Sonar | |
|---|---|---|
| Provider | NVIDIA | Perplexity |
| Noometry Index | 39.6 | 37.8 |
| Released | 2026-03-11 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | 262K | 128K |
| Max output | 131K | 4K |
| Input $ / M tokens | $0.08 | $1 |
| Output $ / M tokens | $0.45 | $1 |
| Results tracked | 21 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Super leads
Nemotron 3 Super: 38.9 (#153), Sonar: 35.1 (#217)
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| SciCode | 36% | — |
| WeirdML | 38% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1403 | — |
| ALE-Bench | 213.9 | — |
Reasoning Sonar leads
Nemotron 3 Super: 17.9 (#274), Sonar: 20.3 (#221)
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| NYT Connections (extended) | 15.4% | — |
| CritPt | 3.1% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1386 | — |
| LiveBench Data Analysis | — | 37.9% |
| LiveBench | — | 46.9% |
Math Nemotron 3 Super leads
Nemotron 3 Super: 37.6 (#95), Sonar: 31.3 (#197)
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| MathArena Final-Answer Competitions | 60.4% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1381 | — |
Knowledge Not comparable
Nemotron 3 Super: 38.2 (#136), Sonar: —
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| LMArena Expert | 1397 | — |
Multilingual Not comparable
Nemotron 3 Super: 48.6 (#139), Sonar: —
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| LMArena Non-English | 1356 | — |
| LMArena Chinese | 1430 | — |
| LMArena French | 1404 | — |
| LMArena German | 1341 | — |
| LMArena Russian | 1336 | — |
| LMArena Spanish | 1417 | — |
Instruction Following Too close to call
Nemotron 3 Super: 71.4 (#149), Sonar: 71.5 (#146)
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| LMArena Instruction Following | 1347 | — |
Long Context Not comparable
Nemotron 3 Super: 41.5 (#134), Sonar: —
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| LMArena Longer Query | 1362 | — |
Writing & Preference Nemotron 3 Super leads
Nemotron 3 Super: 56.1 (#135), Sonar: 52.6 (#166)
| Benchmark | Nemotron 3 Super | Sonar |
|---|---|---|
| LMArena Text | 1378 | — |
| LMArena Creative Writing | 1317 | — |
| LMArena Multi-Turn | 1369 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Nemotron 3 Super better than Sonar?
Nemotron 3 Super is the stronger model overall, scoring 39.6 to 37.8 on the Noometry Index.
Which is cheaper, Nemotron 3 Super or Sonar?
Nemotron 3 Super is cheaper. It lists at $0.08 per million input tokens and $0.45 per million output tokens; Sonar lists at $1 and $1.
Is Nemotron 3 Super or Sonar better for coding?
Nemotron 3 Super scores higher on coding benchmarks: 38.9 versus 35.1 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Super does, with 262K tokens against 128K.
How many benchmarks do Nemotron 3 Super and Sonar share?
0 benchmarks have published results for both models. Nemotron 3 Super has 21 scored results on Noometry and Sonar has 7.