Nemotron 3 Super

Nemotron 3 Super is the middle tier in NVIDIA's Nemotron line, designed for collaborative agents and high-volume reasoning workloads. It was released on 2026-03-11. It has a context window of 262,144 tokens and a maximum output of 262,144 tokens. Weights are open. It supports reasoning mode and tool calling. It accepts text and returns text. Pricing starts at $0.05 per million input tokens and $0.25 per million output tokens, with cached input at $0.025 per million tokens. It is served by 9 hosts.

Against Nemotron Nano 12B v2 VL: window 128K to 262K, input $0.20 to $0.05, output $0.60 to $0.25, cached input $0.04 to $0.025.

Range across the hosts that serve it: up to $0.30 per 1M input tokens.

• ·262k token context window
• ·262k token max output
• ·Supports reasoning mode
• ·Supports tool calling
• ·Open weights for customization

• Reach for it for collaborative agent workflows
• Reach for it for high-volume reasoning tasks
• Reach for it for long-context text generation
• Reach for it for tool-augmented responses

9 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window and output limit?

The context window is 262,144 tokens and the maximum output is also 262,144 tokens.

What is the pricing?

Input starts at $0.05 per million tokens, output at $0.25 per million, and cached input at $0.025 per million.

Does it support reasoning and tool calling?

Yes, it has a reasoning mode and supports tool calling.