Skip to content
NINJA

Nemotron 3 Super

Released by NVIDIA on 11 Mar 2026.

API id nvidia/nemotron-3-super-120b-a12b

Input price
$0.075 / 1M tokens
Output price
$0.40 / 1M tokens
Context window
262K tokens
Longest reply
16K tokens

About this model

NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...

Description from the model’s public catalogue listing.

Capabilities

Reads
Text
Writes
Text
Reasoning mode
Yes
Tool calling
Yes
Open weights
Yes
Cached input
Not listed

What it would cost

At list price, without caching or batch discounts.

1,000 chat replies2K tokens in, 500 out each
$0.35
Summarise 100 long reports50K tokens in, 1K out each
$0.42
A coding agent’s working day20M tokens in, 1M out
$1.90
Price your own workload on every model

Price history

Per 1M tokens, input then output. Tracked since 30 Sept 2026.

6 Nov 2026
$0.30 → $0.075-75%$2.40 → $0.40-83%
28 Oct 2026
$0.08 → $0.30+275%$0.45 → $2.40+433%
28 Oct 2026
$0.30 → $0.08-73%$2.40 → $0.45-81%
21 Oct 2026
$0.08 → $0.30+275%$0.45 → $2.40+433%
21 Oct 2026
$0.30 → $0.08-73%$2.40 → $0.45-81%
21 Oct 2026
$0.08 → $0.30+275%$0.45 → $2.40+433%
8 Oct 2026
$0.085 → $0.08-6%$0.40 → $0.45+13%
8 Oct 2026
$0.08 → $0.085+6%$0.45 → $0.40-11%

Alternatives at a similar price

From other labs, priced within about 40% of Nemotron 3 Super, with a comparable context window.

More from NVIDIA

See all