Model comparison
Llama-3.3-70B-Instruct vs Nemotron 3 Super
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 30.6 on the Noometry Index.
Last verified . 18 shared benchmarks.
Meta
NVIDIASummary
- They share 18 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 0 categories and Nemotron 3 Super in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Nemotron 3 Super leads 39.6 to 15.3.
- The biggest single-benchmark swing is WeirdML: 14.4% for Llama-3.3-70B-Instruct and 38% for Nemotron 3 Super.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.08 / $0.45 for Nemotron 3 Super.
- Nemotron 3 Super accepts more context: 262K tokens versus 128K.
Side by side
| Llama-3.3-70B-Instruct | Nemotron 3 Super | |
|---|---|---|
| Provider | Meta | NVIDIA |
| Noometry Index | 30.6 | 40.1 |
| Released | 2024-12-06 | 2026-03-11 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.10 | $0.08 |
| Output $ / M tokens | $0.32 | $0.45 |
| Results tracked | 43 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Super leads
Llama-3.3-70B-Instruct: 31.0 (#290), Nemotron 3 Super: 39.4 (#158)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| SciCode | 26% | 36% |
| WeirdML | 14.4% | 38% |
| LMArena Coding | 1268 | 1403 |
| BigCodeBench Instruct | 46.9% | — |
| LiveBench Coding | 36.6% | — |
| BigCodeBench Complete | 57.5% | — |
| ALE-Bench | — | 213.9 |
Agentic & Tool Use Not comparable
Llama-3.3-70B-Instruct: 25.8 (#105), Nemotron 3 Super: —
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | — |
| BALROG | 23% | — |
Reasoning Nemotron 3 Super leads
Llama-3.3-70B-Instruct: 14.1 (#327), Nemotron 3 Super: 18.7 (#276)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| CritPt | 0% | 3.1% |
| LMArena Hard Prompts | 1257 | 1386 |
| SimpleBench | 19.9% | — |
| NYT Connections (extended) | — | 15.4% |
| LiveBench Reasoning | 50.8% | — |
| DTBench | 59.5% | — |
| LiveBench Data Analysis | 49.5% | — |
| LMCA | 17.5% | — |
| Epoch Capabilities Index | 127.33 | — |
| ForecastBench | 58.6 | — |
| LiveBench | 50.2% | — |
Math Nemotron 3 Super leads
Llama-3.3-70B-Instruct: 15.3 (#298), Nemotron 3 Super: 39.6 (#101)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| LMArena Math | 1267 | 1380 |
| MathArena Final-Answer Competitions | — | 60.4% |
| OTIS Mock AIME 2024-2025 | 5.1% | — |
| LiveBench Math | 42.2% | — |
| MATH Level 5 | 41.6% | — |
Knowledge Nemotron 3 Super leads
Llama-3.3-70B-Instruct: 30.6 (#226), Nemotron 3 Super: 38.9 (#139)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| LMArena Expert | 1225 | 1398 |
| GPQA Diamond | 47.4% | — |
| Confabulations | 22.8% | — |
| Vectara Hallucination Rate | 4.1% | — |
| MMLU | 86.3% | — |
Multilingual Nemotron 3 Super leads
Llama-3.3-70B-Instruct: 39.9 (#220), Nemotron 3 Super: 48.4 (#144)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| LMArena Non-English | 1236 | 1355 |
| LMArena Chinese | 1217 | 1432 |
| LMArena French | 1281 | 1405 |
| LMArena German | 1251 | 1341 |
| LMArena Russian | 1252 | 1336 |
| LMArena Spanish | 1270 | 1417 |
| LMArena Japanese | 1150 | — |
| LMArena Korean | 1143 | — |
Instruction Following Too close to call
Llama-3.3-70B-Instruct: 71.1 (#157), Nemotron 3 Super: 71.2 (#154)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| LMArena Instruction Following | 1242 | 1347 |
| LiveBench Instruction Following | 82.7% | — |
Long Context Nemotron 3 Super leads
Llama-3.3-70B-Instruct: 26.4 (#295), Nemotron 3 Super: 41.5 (#139)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| LMArena Longer Query | 1256 | 1362 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Nemotron 3 Super leads
Llama-3.3-70B-Instruct: 47.6 (#207), Nemotron 3 Super: 55.9 (#140)
| Benchmark | Llama-3.3-70B-Instruct | Nemotron 3 Super |
|---|---|---|
| LMArena Text | 1274 | 1378 |
| LMArena Creative Writing | 1250 | 1317 |
| LMArena Multi-Turn | 1280 | 1369 |
| LiveBench Language | 39.2% | — |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than Nemotron 3 Super?
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 30.6 on the Noometry Index.
Which is cheaper, Llama-3.3-70B-Instruct or Nemotron 3 Super?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Nemotron 3 Super lists at $0.08 and $0.45.
Is Llama-3.3-70B-Instruct or Nemotron 3 Super better for coding?
Nemotron 3 Super scores higher on coding benchmarks: 39.4 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Super does, with 262K tokens against 128K.
How many benchmarks do Llama-3.3-70B-Instruct and Nemotron 3 Super share?
18 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Nemotron 3 Super has 21.