Skip to content
Dashboard

NVIDIA Nemotron 3 Super 120B A12B

NVIDIA Nemotron 3 Super 120B A12B is NVIDIA's 120B total, 12B active-parameter hybrid Mamba-Transformer MoE built for complex multi-agent applications, featuring latent MoE and multi-token prediction.

ReasoningTool Use
index.ts
import { streamText } from 'ai'
const result = streamText({
model: 'nvidia/nemotron-3-super-120b-a12b',
prompt: 'Why is the sky blue?'
})

More models by NVIDIA

Model
Context
Latency
Throughput
Input
Output
Cache
Web Search
Capabilities
Providers
ZDR
No Training
Release Date
1M
0.3s
234tps
$0.37/M
$1.08/M
Read:$0.12/M
Write:
baseten logo
blackbox logo
deepinfra logo
+1
06/04/2026
262K
0.3s
83tps
$0.05/M
$0.24/M
deepinfra logo
12/15/2025
131K
0.2s
199tps
$0.20/M
$0.60/M
bedrock logo
deepinfra logo
10/28/2025
131K
0.2s
176tps
$0.06/M
$0.23/M
bedrock logo
deepinfra logo
08/18/2025