Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. NVIDIA

Nvidia Nemotron 3 Super 120B

Nvidia's Nemotron 3 Super 120B A12B model from the March 2026 Nemotron 3 release. It uses a hybrid Mamba-Transformer MoE architecture and targets agentic and coding workloads with a 262K context window here.

  • Reasoning

Added Mar 1, 2026

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.05
Output
$0.25
Cache read
$0.025
Compare provider prices

Specifications

Context window
262.1K
Max output
16.4K
Parameters
120B / 12B
Total / active

Benchmarks

Sourced from LMArena.

Arena Score

1359.7

Overall Rank

#212 / 421

Votes

7,842

Confidence Interval

1352.5 - 1366.8

Category Scores

Coding

#211 / 416

1,877 votes

1407.1

Math

#201 / 404

544 votes

1370.1

Longer Query

#220 / 399

2,265 votes

1357.2

Creative Writing

#242 / 419

1,118 votes

1301.6

Instruction Following

#220 / 421

2,151 votes

1342.3

Hard Prompts

#211 / 421

4,380 votes

1377.7

Additional Categories
20

Spanish

#167 / 304

353 votes

1388.1

Industry Mathematical

#180 / 402

443 votes

1397.3

French

#185 / 301

277 votes

1385.9

Chinese

#190 / 395

583 votes

1402.0

Expert

#194 / 371

657 votes

1394.0

English

#205 / 421

3,134 votes

1383.2

Industry Software And It Services

#207 / 421

2,823 votes

1402.6

German

#209 / 320

200 votes

1313.9

Polish

#211 / 243

275 votes

1309.6

Exclude Ties

#213 / 421

5,383 votes

1321.7

Hard Prompts English

#215 / 419

1,758 votes

1390.2

Multi Turn

#217 / 419

1,355 votes

1349.1

Non English

#217 / 421

4,708 votes

1336.5

Industry Legal And Government

#219 / 393

551 votes

1366.0

Industry Business And Management And Financial Operations

#221 / 414

1,454 votes

1350.2

Industry Entertainment And Sports And Media

#222 / 419

1,253 votes

1314.6

Industry Life And Physical And Social Science

#222 / 419

1,221 votes

1367.6

Industry Writing And Literature And Language

#229 / 420

1,723 votes

1323.6

Industry Medicine And Healthcare

#237 / 391

507 votes

1350.2

Russian

#238 / 385

1,020 votes

1325.4

Published 2026-11-11 · Matched as NVIDIA-Nemotron-3-Super-120B-A12B · CC BY 4.0

LMArena Dataset

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Auto
Auto (ZDR)
$0.05$0.25$0.025N/AN/A
Auto
Auto (Anonymized)
$0.05$0.25$0.025N/AN/A
Ollama
Ollama
UNK
$0.0158$0.63$0.0158N/AN/A
DeepInfra
DeepInfra
BF16
US
$0.0787$0.42N/AN/AN/A
DekaLLM
DekaLLM
FP8
ID
$0.084$0.472N/AN/AN/A
Uomi
Uomi
GLB
$0.0992$0.496N/AN/AN/A
Privacy:ZDR, prompts not retainedNo training on prompts
Batch API
ProviderInputOutputCached input
Auto
Autonvidia/nemotron-3-super-120b-a12b
$0.05/M$0.25/M$0.025/M

Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.

Related text models

Compare Nvidia Nemotron 3 Super 120B with similar models from the same provider or model family.

Nvidia Nemotron 3 Super 120B Thinking

nvidia/nemotron-3-super-120b-a12b:thinking

Nvidia Nemotron 3 Super 120B with reasoning content enabled. Returns separate thinking content when requested.

Nvidia Nemotron 3.5 Super VL

nvidia/nemotron-3.5-super-vl

NVIDIA’s open-weight 120B model (12B active) for reasoning, coding, tool use, document and GUI understanding, and video understanding. Text, images and video in; text out. A 262,144-token shared context window with responses limited to 16,384 tokens. OpenMDW 1.1 license.

Nvidia Nemotron Super 49B

nvidia/Llama-3.3-Nemotron-Super-49B-v1

Llama-3.3-Nemotron-Super-49B-v1 is a model which offers a great tradeoff between model accuracy and efficiency. Efficiency (throughput) directly translates to savings. Using a novel Neural Architecture Search (NAS) approach, we greatly reduce the model's memory footprint, enabling larger workloads, as well as fitting the model on a single GPU at high workloads (H200). This NAS approach enables the selection of a desired point in the accuracy-efficiency tradeoff. For more information on the NAS approach, please refer to this paper.

Nemotron 3.5 Content Safety

nvidia/nemotron-3.5-content-safety

Nemotron 3.5 Content Safety is a content moderation classifier that labels user messages and assistant responses as safe or unsafe. It supports a 131,072-token context window and optional reasoning.

Nvidia Nemotron 3.5 Lightning

nvidia/nemotron-3.5-lightning

Nvidia Nemotron 3.5 Lightning is an open-weight 30B-A3B hybrid Mamba-Transformer mixture-of-experts model for high-throughput agentic, coding, tool-use, instruction-following, and long-context workloads. Thinking is disabled on this variant.

Nvidia Nemotron 3.5 Lightning Thinking

nvidia/nemotron-3.5-lightning:thinking

Nvidia Nemotron 3.5 Lightning is an open-weight 30B-A3B hybrid Mamba-Transformer mixture-of-experts model for high-throughput agentic, coding, tool-use, instruction-following, and long-context workloads. This variant enables its reasoning trace.