Back
FR

Model creator

NVIDIA

Every NVIDIA model in the BenchSift catalogue, with its scores, ranks, prices and capabilities as the sources publish them.

Hugging Face · nvidia
View in the catalogue
Models
45
53 configurations, reasoning levels included
Open weights
40
40 of 45 models
Latest release
Oct 29, 2026
Nemotron 3.5 Transcribe · First dated model: Aug 21, 2024
Median price
$0.30
$0.086 to $1.20 per million tokens, across 11 priced models

Best scores and ranks

The creator's best model on each measure, and its rank among every measured model in the catalogue.

Intelligence

Nemotron 3.5 Bolt 74B-A9B
21.4Rank 242 of 782 measured models

Coding

Nemotron 3 Ultra 550B A55B (Reasoning)
49.3Rank 126 of 301 measured models

Math

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)
91.0Rank 21 of 270 measured models

Agentic

Nemotron 3 Ultra 550B A55B (Reasoning)
20.1Rank 99 of 191 measured models

Output speed

Nemotron 3.5 Super VL
263 t/sRank 16 of 449 measured models

Lowest price

NVIDIA Nemotron Nano 9B V2 (Reasoning)
$0.070Rank 74 of 740 measured models

OpenRouter ranking

Nemotron 3.5 Super VL
#30

Capabilities

Number of models that declare each capability. Tool calling and structured outputs come from OpenRouter.
Reasoning25 of 45
Open weights40 of 45
Tool calling2 of 45
Structured outputs2 of 45

Input

  • Text40
  • Image12
  • Audio5
  • Video4

Output

  • Text35
  • Image4
  • Speech4
  • Transcription3
  • Video2

Overview

Model types

  • Text35
  • Audio4
  • Image4
  • Video2

Releases by year

  • 202618
  • 202518
  • 20242

Hugging Face

Official repositories confirmed by Hugging Face metadata.

2 official repositories

Licenses

  • other2

Models

45

Reasoning levels are folded into one entry. The rank is the whole catalogue's: sorting does not change it.

Where these figures come from

Each catalogue model is tied to its creator by the identity Artificial Analysis publishes, or OpenRouter's and Models.dev's for models it does not track. Counts fold the reasoning levels of one model together. Ranks compare each measure with every measured model in the catalogue, configurations included, as model pages do. The price is the published blended price (3 input tokens to 1 output token).

Figures computed by BenchSift from the catalogue: Artificial Analysis, OpenRouter, Models.dev and Hugging Face. A value no model carries is not shown.

Data retrieved on Nov 12, 2026.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy