nemotron-cascade-2:30b

1,820 Downloads Updated 15 hours ago

An open 30B MoE model from NVIDIA with 3B activated parameters that delivers strong reasoning and agentic capabilities.

tools thinking 30b

ollama run nemotron-cascade-2:30b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "nemotron-cascade-2:30b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='nemotron-cascade-2:30b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'nemotron-cascade-2:30b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 17 hours ago

17 hours ago

e0705e3fe8f7 · 24GB ·

model

archnemotron_h_moe

parameters31.6B

quantizationQ4_K_M

24GB

license

NVIDIA Open Model License Agreement Last Modified: October 24, 2025 This NVIDIA Open Model License A

10kB

params

{ "temperature": 1, "top_p": 0.95 }

31B

Readme

Nemotron-Cascade-2-30B-A3B is an open 30B MoE model with 3B activated parameters that delivers strong reasoning and agentic capabilities. It achieves gold medal performance in both the 2025 International Mathematical Olympiad (IMO) and the International Olympiad in Informatics (IOI). It operates in both thinking and instruct (non-reasoning) modes.