Nemotron Cascade 2 — local AI model by NVIDIA
Written by Jakub Rusinowski · Last updated
NVIDIA's hybrid Mamba-2 + MoE model, released March 2026 as a single checkpoint: Nemotron-Cascade-2-30B-A3B. 31.6B parameters resident, 3.5B active per token, and the second open-weight model to reach gold-medal level at IMO 2025, IOI and the ICPC World Finals. NVIDIA also open-sourced the post-training recipe. NVIDIA Open Model License, 1M-token context, first-party Ollama tag.
Variants
The smallest Nemotron Cascade 2 variant needs about 20 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| Nemotron-Cascade 2 30B-A3B → 31.6B (3.5B active) | ~19.9 GB | 1,048,576 | ollama run nemotron-cascade-2:30b | |
| Nemotron Cascade 2 70B 70B | ~43.1 GB | 128,000 | ollama run nemotron-cascade-2 |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run Nemotron Cascade 2 locally
Install Ollama, then pull the tag.
ollama run nemotron-cascade-2:30bPick a size above for its own VRAM figure, speed estimate and install command.
Licence
Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
Applies to: Nemotron-Cascade 2 30B-A3BCommercial use permitted. No usage restrictions beyond attribution.
Applies to: Nemotron Cascade 2 70BRecommended GPU
The cheapest catalogued GPU that runs Nemotron Cascade 2 locally (min 20 GB VRAM) is the AMD Radeon RX 7900 XT (20 GB).
Can I run Nemotron Cascade 2 on my GPU?
- Nemotron Cascade 2 on AMD Radeon RX 7800 XT
- Nemotron Cascade 2 on AMD Radeon RX 7900 XT
- Nemotron Cascade 2 on AMD Radeon RX 7900 XTX
- Nemotron Cascade 2 on AMD Radeon RX 9060 XT 16GB
- Nemotron Cascade 2 on AMD Radeon RX 9060 XT 8GB
- Nemotron Cascade 2 on AMD Radeon RX 9070
- Nemotron Cascade 2 on AMD Radeon RX 9070 XT
- Nemotron Cascade 2 on AMD Ryzen AI Max+ 395
- Nemotron Cascade 2 on Apple M1
- Nemotron Cascade 2 on Apple M1 Max
- Nemotron Cascade 2 on Apple M1 Pro
- Nemotron Cascade 2 on Apple M2
- Nemotron Cascade 2 on Apple M2 Max
- Nemotron Cascade 2 on Apple M2 Pro
- Nemotron Cascade 2 on Apple M3
- Nemotron Cascade 2 on Apple M3 Max
- Nemotron Cascade 2 on Apple M3 Pro
- Nemotron Cascade 2 on Apple M4
- Nemotron Cascade 2 on Apple M4 Pro
- Nemotron Cascade 2 on Apple M5
- Nemotron Cascade 2 on Apple M5 Pro
- Nemotron Cascade 2 on Intel Arc B570
- Nemotron Cascade 2 on Intel Arc B580
- Nemotron Cascade 2 on NVIDIA A100 80GB (PCIe)
- Nemotron Cascade 2 on NVIDIA GeForce RTX 3060 (12GB)
- Nemotron Cascade 2 on NVIDIA GeForce RTX 3070
- Nemotron Cascade 2 on NVIDIA GeForce RTX 3070 Ti
- Nemotron Cascade 2 on NVIDIA GeForce RTX 3080 (10GB)
- Nemotron Cascade 2 on NVIDIA GeForce RTX 3090
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4060
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4060 Ti 16GB
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4070
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4070 Super
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4070 Ti
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4070 Ti Super
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4080
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4080 Super
- Nemotron Cascade 2 on NVIDIA GeForce RTX 4090
- Nemotron Cascade 2 on NVIDIA GeForce RTX 5060
- Nemotron Cascade 2 on NVIDIA GeForce RTX 5060 Ti 16GB
- Nemotron Cascade 2 on NVIDIA GeForce RTX 5060 Ti 8GB
- Nemotron Cascade 2 on NVIDIA GeForce RTX 5070
- Nemotron Cascade 2 on NVIDIA GeForce RTX 5070 Ti
- Nemotron Cascade 2 on NVIDIA GeForce RTX 5080
- Nemotron Cascade 2 on NVIDIA GeForce RTX 5090
- Nemotron Cascade 2 on NVIDIA H100 80GB (PCIe)
- Nemotron Cascade 2 on NVIDIA L40S
- Nemotron Cascade 2 on NVIDIA RTX 6000 Ada Generation
- Nemotron Cascade 2 on NVIDIA RTX PRO 6000 Blackwell
Nemotron Cascade 2 — frequently asked questions
How much VRAM does Nemotron Cascade 2 need?
Nemotron Cascade 2 needs about 20 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Nemotron-Cascade 2 30B-A3B (20 GB, Q4_K_M); Nemotron Cascade 2 70B (43 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Nemotron Cascade 2 on an RTX 4090 (24 GB)?
Yes — Nemotron Cascade 2 runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for Nemotron Cascade 2?
Q4_K_M is the best balance of quality and VRAM for Nemotron Cascade 2 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Nemotron Cascade 2 with Ollama?
Install Ollama, then run: ollama run nemotron-cascade-2:30b. This downloads Nemotron Cascade 2 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.