Nemotron Cascade 2 30B-A3B

Overview

NVIDIA's 30B MoE (3B active) with both thinking and instruct modes. Earned IMO 2025 and IOI 2025 gold medals — 30B-class reasoning at 3B-active inference speed. Released April 2026.

When to pick this model

• Competition-grade math and code workloads
• Reasoning agents needing fast inference (3B active)
• Single-GPU deployments on 24 GB cards in Q4
• Production systems on NVIDIA Open Model License terms
• Tasks switching between thinking and instruct modes

VRAM requirements by quantization

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

Published benchmark scores

Scores published by the model author or aggregated from public leaderboards. Re-measured monthly by our editorial team.

Strengths

• Gold medal at IMO 2025 and IOI 2025 in thinking mode
• Fast inference with only 3B active params
• Fits on a 24 GB GPU at Q4
• Commercial use allowed under NVIDIA Open Model License

Limitations

• NVIDIA Open Model License — not Apache or MIT
• 32+ GB VRAM total in Q4 (full model is 30B)
• Thinking mode generation can be slow

Architecture & training

Architecture: MoE 30B/3B active · unified thinking mode + instruct · 128k ctx

Training: Trained by NVIDIA. Gold medal at IMO 2025 and IOI 2025 in thinking mode. Optimized for mathematical reasoning and competitive code.

Quick start

ollama run nemotron-cascade-2

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.