- AIME25 (reasoning on) 78.5
- MATH500 (reasoning on) 95.4
- GPQA (reasoning on) 53.2
- LCB (reasoning on, version not stated) 51.8
- BFCL v3 61.1
- IFEVAL-Prompt (reasoning on); Instruction 92 87.9
- Mean of Tau2 Airline 33.3 / Retail 39.8 / Telecom 33 (reasoning on) 35.4
- IFBench-Prompt (reasoning off); Instruction 44.2 43.2
- Orak (reasoning off) 22.9
- HaluEval (reasoning off) 62.2
- RULER (128k, reasoning off) 91.1
- EQ-Bench3 (reasoning off) 63.2
Nemotron 3 Nano 4B
Nemotron 3 Nano 4B.
13
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
NVIDIA Mar 2026 Nemotron 3 Nano 4B
Open weights Small
No Epoch Capabilities Index score yet
- IFBench 43.2% #66 /70
- AgentIF (2026-10 setting) 58.5% #18 /24
- tau2-bench 35.4% #69 /69
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| IFEval | 87.9% | #48 of 126 | Model card | |
| RULER | 91.1% | #18 of 38 | Model card | |
| AgentIF (2026-10 setting) | 58.5% | #18 of 24 | Official leaderboard | |
| IFBench | 43.2% | #66 of 70 | Model card | |
| EQ-Bench 3 | 588 | #77 of 78 | Official leaderboard | |
| HaluEval | 62.2% | #1 of 1 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MATH-500 | 95.4% | #39 of 67 | Model card | |
| AIME 2025 | 78.5% | #90 of 143 | Model card | |
| GPQA Diamond | 53.2% | #224 of 284 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench | 51.8% | #42 of 73 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Berkeley Function Calling Leaderboard (BFCL) | 61.1% | #40 of 47 | Model card | |
| tau2-bench | 35.4% | #69 of 69 | Model card | |
| Orak | 22.9 | #1 of 1 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.