NOUMENOpen app
NOUMEN

nvidia · Open weights

Nemotron 3.5 Content Safety

Added Sep 19, 2026Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

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.

  • Reasoning
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.05
Output
$0.15
Cache read
$0.025

Reference prices as published by the model provider. API example

Specifications

Context window
131.1K
Max output
32.8K
Average output (7 days)
16 tokens
Longer than 0% of 465 models

Side by side

Nemotron 3.5 Content Safety compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Nemotron 3.5 Content SafetynvidiaThis modelNvidia Nemotron 3 Ultra 550B ThinkingnvidiaRemove
Providernvidianvidia
Discovery tierOpen weightsOpen weights
AddedSep 19, 2026Jun 4, 2026
Limits
Context window131.07K1MBest
Max output32.77K65.54KBest
Average output16 tokens654 tokens
Pricing, USD per 1M tokens
Input$0.05Best$0.50
Output$0.15Best$2.50
Cache read$0.025—
Performance
Throughput—79.9 tok/s
Time to first token—2.2 s
Request success (24h)—99.7%
Benchmark indexes
Intelligence Index—22.9
Coding Index—49.3
Capabilities
ReasoningYesYes
ToolsNoYes
Parallel toolsNoYes
Reasoning effort——

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Nemotron 3.5 Content Safety
  • Nvidia Nemotron 3 Ultra 550B Thinking

Intelligence Index

  • Nemotron 3.5 Content Safety—
  • Nvidia Nemotron 3 Ultra 550B Thinking22.9
3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 49.3

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 3.0%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 0.3%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 876 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 1,013 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 5.0%

  • AA-LCRv1.1

    Reasoning across long documents

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 79.3%

  • MLCR-AA

    Long-context reasoning over medical records

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 5.6%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 28.4%

  • GPQA Diamond

    Graduate-level science questions

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 86.7%

  • IFBench

    Following precise output instructions

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 81.4%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 0.5%

  • SciCode

    Scientific computing in Python

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 40.3%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 36.4%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 83.3%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Nemotron 3.5 Content Safety: —
    • Nvidia Nemotron 3 Ultra 550B Thinking: 79.3%

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/nemotron-3.5-content-safety",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="nvidia/nemotron-3.5-content-safety",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

nvidia

Other nvidia models, starting with the same family.