τ²-bench
- Other labs
- Reported in a later post or card
Results
| Model | Score | Source | Source check |
|---|---|---|---|
| Nemotron 3 Ultra 550B-A55B4 Jun 2026 · NVIDIA | 79.75% | Nemotron 3.5 Super and Lightning: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Models for Agentic AIaverage of 8 trials; Qwen3-235B-A22B-Instruct-2507 user simulator; reported later | 1 other sourceThe same number is printed in 1 other lab document, linked below. |
| Nemotron 3 Super 120B-A12B11 Mar 2026 · NVIDIA | 56.25% | nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 model card (launch-day revision)default prompt | 3 other sourcesThe same number is printed in 3 other lab documents, linked below. |
| Nemotron 3 Nano 30B-A3B15 Dec 2025 · NVIDIA | 48.0% | nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8 model card (launch-day revision) | 3 other sourcesThe same number is printed in 3 other lab documents, linked below. |
Benchmark
github.comHigher is better
SierraIntroduced 9 Jun 2025