nemotron-cascade-2:30b-a3b

155.5K 2 hours ago

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

tools thinking 30b
... /
json
f34aa5ef920a ยท 138B
{
"_from_model_config": true,
"bos_token_id": 1,
"eos_token_id": [2, 11],
"pad_token_id": 0,
"transformers_version": "4.55.4"
}