nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 · Hugging Face

Source: https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#descriptionDescription

NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16is the full-precision release of NVIDIA Nemotron 3.5 Super VL, a vision-language model for reasoning, coding, tool use, document understanding, and video understanding. The BF16 weights support customization, domain adaptation, and full-precision evaluation. For optimized deployment, see theNVFP4 release.

The model combines interleaved Mamba-2, Mixture-of-Experts (MoE), and attention layers with a C-RADIO vision encoder and a vision-language connector. It has120B total parametersand12B active parametersand supports context lengths up to1M tokens. Multi-Token Prediction (MTP) enables native speculative decoding.

This model is ready for commercial or non-commercial use.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#model-summaryModel Summary

Total Parameters120B (12B active)ArchitectureMamba-2–Transformer hybrid Mixture of Experts with a C-RADIO vision encoderPrecisionBF16 (full-precision reference weights)Context LengthUp to 1M tokens; deployment settings are listed belowInputs / OutputsText, images, and video / textDeployment4× H100 or 2× GB200Speculative DecodingNative Multi-Token Prediction (MTP)Reasoning ModeConfigurable through the chat template (enable\_thinking=True/False)Recommended SamplingText: temperature 1.0, top_p 0.95; vision: temperature 0.6, top_p 0.95Best ForCustomization, domain adaptation, and full-precision evaluationLicenseOpenMDW-1.1

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#model-overviewModel Overview

**Model Developer:**NVIDIA Corporation

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#what-is-nemotronWhat is Nemotron?

NVIDIA Nemotron™ is a family of open models with open weights, training data, and recipes, delivering leading efficiency and accuracy for building specialized AI agents.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#quick-startQuick Start

For efficient inference on DGX Spark or data center GPUs, see theNVFP4 card. The example below serves BF16 on4× H100.

Use Docker with NVIDIA GPU support and thevllm/vllm\-openai:nightly\-3f1bcfd82bb2e93fcd364fc7a5491718453a77e2or (vllm/vllm\-openai:v0\.32\.0) image. If the model requires authentication, setHF\_TOKENin your shell before launching the container.

export MODEL=nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16

vllm serve $MODEL \
  --served-model-name nemotron-3.5-super-vl \
  --tensor-parallel-size 4 \
  --trust-remote-code \
  --gpu-memory-utilization=0.85 \
  --allowed-local-media-path=/ \
  --max-num-seqs=32 \
  --limit-mm-per-prompt '{"video": 1, "image": 1}' \
  --video-pruning-rate 0.5 \
  --media-io-kwargs '{"video": {"fps": 2, "num_frames": 256}}' \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder \
  --reasoning-parser nemotron_v3 \
  --speculative_config.method mtp \
  --speculative_config.num_speculative_tokens 3

Wait for the server to finish loading the model and compiling kernels, then check the endpoint:

curl --fail http://localhost:8000/v1/models

All vLLM and SGLang examples use port8000and the API model namenemotron\-3\.5\-super\-vl. See theAPI Clientfor text, image, video, and tool-calling examples.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#licenseterms-of-useLicense/Terms of Use

**Governing Download Terms:**Use of this model is governed by theOpenMDW License Agreement, version 1.1.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#benchmarksBenchmarks

We evaluated our model on the following benchmarks:

BenchmarkNemotron 3.5 SuperGA BF16Qwen3.5-122B-A10BQwen3.8-Flash-NextDeepSeek-V4-Flash0731Nemotron 3 UltraInkling-SmallGLM-5.3-FlashGPQA-Diamond89.5286.792.789.687.088.691.2HLE, text-only31.7924.239.635.829.233.336.47MMLU-ProX76.9183.283.085.583.184.083.0AA-Omniscience accuracy26.2026.227.945.025.335.434.2AA-Omniscience hallucination rate30.084.541.395.922.159.630.3AA-Omniscience index4.03-36.1-1.9-7.88.8-3.114.3CritPt10.861.1—12.63.19.1—APEX shortlist88.3654.791.888.775.071.577.1LiveCodeBench v688.2276.794.194.689.288.380.6SciCode, subtask47.1339.846.649.342.048.350.05AA-LCR v1.168.6275.178.479.775.277.379.5GraphWalks89.7186.176.165.463.355.788.4IFBench70.8667.675.676.282.979.357.7LMArena v3 proxy52.6747.035.249.834.353.661.0WMT24++84.8086.386.285.582.784.686.1tau2-bench retail69.2455.471.673.873.771.878.2tau2-bench airline76.5063.580.883.579.862.781.8tau2-bench telecom73.1987.290.498.283.090.198.5tau3-bench banking18.8814.244.538.618.4—34.8GDPval-AA v2121186115611382114813131563BrowseComp7858.189.084.267.8574.186.6SWE-bench Verified71.1071.978.775.473.975.2779.10SWE-bench Multilingual67.7262.077.1072.068.871.478.30SWE-bench Pro51.7844.960.3756.9149.6656.5459.70Terminal-Bench 2.166.7048.7085.8076.8054.8053.4082.70BabyVision29.6035.746.8N/AN/A24.739.1MMLongBench-Doc62.8263.767.9N/AN/A49.0665.8V*85.6091.193.0N/AN/A70.588.1OCR-Reasoning65.0864.367.1N/AN/A54.764.6MathVision78.1683.387.5N/AN/A78.882.5HLE, full set29.2024.438.8N/AN/A31.635.1OCRBench v2, EN68.1268.367.9N/AN/A54.261.8OCRBench v2, ZH59.1567.368.3N/AN/A52.456.8CharXiv reasoning80.8283.585.8N/AN/A77.881.9Video-MME v232.8532.045.1N/AN/A—14.8OSWorld-Verified51.1451.555.8N/AN/A—14.78BrowseComp [offline]7858898344 (reported)7486.6WebVoyager [offline]77.7254.8990.39not applicablenot applicable5.2572.64AA-Briefcase-Lite [offline]1137 (-) | 1252 (-) | 40551 | 591 | 221636 | 1324 | 761469 | 1277 | 80665 (836) | 866 (874) | 47765 | 684 | 171785 (1658) | 1428 (1408) | 76AutomationBench-AA [offline]32.4427.2572.0663.8623.2832.8370.17GDP.pdf [offline]6.6 (AA 6.6)8.8 (AA 3.4)17.4 (AA 15.6)10.8 (AA 11.0)8.4 (AA 5.0)8.6 (AA 11.2)14.8 (AA 15.4)Terminal Bench 4 [offline]1020.713.5400.515.15DeepSWE [offline]24.650.455.5742.331857.77 / 63.4 (reported) We evaluated Nemotron 3.5 Super on a comprehensive suite of benchmarks spanning agentic capabilities, reasoning and knowledge, multimodal, conversational ability and instruction following, long-context understanding, and multilingual performance. All evaluation results for Nemotron 3.5 Super and baselines were collected via NeMo Gym. For reproducibility purposes, more details on the evaluation settings can be found in theNeMo Gym recipes folderand thecorresponding reproducibility tutorial. The published recipes specify the benchmark-specific containers, prompts, inference parameters, parser configurations, and scoring settings used to produce the results.

There were a few benchmarks that were not yet onboarded in our open source tools, and for these we used either their official open source implementation or an internal scaffolding that we plan to open source in the future: BabyVision, CharXiv-RQ, MathVision, MMLongBench-Doc, OCR-Reasoning, OCRBench-v2, V*, Video-MME-v2.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#agentic-coding-benchmarksAgentic Coding Benchmarks

Additional harness-level coding-agent results for SWE-Bench Pro and Terminal-Bench 2.1 are shown below.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#deployment-geography-globalDeployment Geography: Global

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#use-caseUse Case

This model is intended for developers who want to customize a post-trained vision-language model, adapt it to domain-specific data, develop quantized variants, or run full-precision inference and evaluation.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#release-dateRelease Date

Hugging Face - 10/15/2026 viaHugging Face

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#referencesReference(s)

• NVIDIA Nemotron 3.5 Super VL and Lightning Technical Report

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#model-architectureModel Architecture

• **Architecture Type:**Mamba-2–Transformer hybrid Mixture of Experts (MoE) with Multi-Token Prediction (MTP)
• **Network Architecture:**Nemotron Hybrid MoE with 512 routed experts per MoE layer, a C-RADIO vision encoder, and a vision-language connector
• **Number of Model Parameters:**120B total / 12B active

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#model-designModel Design

The model combines sparse expert computation, attention, and Mamba-2 layers to support long-context processing. The vision encoder and connector turn visual inputs into representations processed by the language backbone. A shared MTP head predicts future tokens recursively for speculative decoding.

The model supports up to 1M tokens of context. The prompt, visual inputs, conversation history, tool results, and generated response share this context window. Individual deployment recipes may use a smaller context window to accommodate the available GPU memory and concurrency.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#training-methodologyTraining Methodology

Stage 1: Foundation and Multi-Token Prediction

The language backbone is initialized from Nemotron 3 Super Base. A shared MTP head is trained for 100B tokens while the backbone remains frozen, followed by 1T tokens of distillation from Nemotron 3 Ultra Base.

Stage 2: Continued Pre-Training and Mid-Training

The model undergoes 100B tokens of Indic continued pre-training, a 35B-token extension to 1M context, 67B tokens of software-engineering mid-training at 256K context, and a final 9B-token long-context recovery stage. Software-engineering data includes repositories, issue–pull-request pairs, commit diffs, build and CI logs, and agent trajectories.

Stage 3: Vision-Language Training

A C-RADIO vision encoder and connector extend the language backbone to visual inputs, including images, document pages, and video frames.

Stage 4: Supervised Fine-Tuning

A two-stage supervised fine-tuning pipeline jointly trains the language and vision components on vision, reasoning, code, tools, chat, safety, and long-context data. Training uses packed sequences of up to 524,288 tokens.

Stage 5: Reinforcement Learning and Distillation

Reinforcement learning with verifiable rewards, reinforcement learning from human feedback, and specialized teacher training improve reasoning, agentic, chat, and vision capabilities. Multi-teacher on-policy distillation consolidates these capabilities into a single model.

Stage 6: MTP Boosting

The MTP head is further trained against the frozen post-trained backbone to improve speculative-decoding acceptance.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#inputInput

• **Input Type(s):**Text, images, and video
• **Input Format(s):**Text strings, images, and sampled video frames- Video: mp4, up to 2 minutes. - Image: Red, Green, Blue (RGB) (jpeg, png)
• Input Parameters:- Video: Three-Dimensional (3D) - Image: Two-Dimensional (2D) - Text: One-Dimensional (1D)
• **Other Properties Related to Input:**Maximum context length up to 1M tokens. Documents can be supplied as page images with accompanying text.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#outputOutput

• **Output Type(s):**Text
• **Output Format:**String
• **Output Parameters:**One-dimensional sequences
• **Other Properties Related to Output:**Maximum context length up to 1M tokens.

Our AI models are designed and optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#software-integrationSoftware Integration

• **Runtime Engine(s):**vLLM; SGLang
• Hardware:- NVIDIA Blackwell - NVIDIA Hopper
• **Operating System(s):**Linux

The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#model-versionsModel Version(s)

• v1.0 - GA

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#quick-start-guideQuick Start Guide

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#choose-your-deploymentChoose your deployment

HardwareRuntimeConfiguration4× H100vLLMText, image, and video; TP4; MTP with 3 draft tokens2× GB200vLLMText, image, and video; TP2; MTP with 5 draft tokens

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#speculative-decodingSpeculative Decoding

The serving examples enable the model’s native MTP head. No separate drafter download is required for these vLLM and SGLang recipes. In vLLM,num\_speculative\_tokenscontrols draft length; in SGLang, the MTP recipe usesEAGLEwithspeculative\-eagle\-topk 1.

Performance depends on hardware, request length, and concurrency. To compare with standard decoding, remove the speculative configuration while keeping the model, workload, sampling parameters, and other server settings the same.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#vllmvLLM

Usevllm/vllm\-openai:nightly\-3f1bcfd82bb2e93fcd364fc7a5491718453a77e2for these commands (orvllm/vllm\-openai:v0\.32\.0).

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#4%C3%97-h100-up-to-1m-context-length4× H100 (up to 1M Context Length)

Use this command if you need to serve the model with up to 1M context length.

vllm serve $MODEL \
  --served-model-name nemotron-3.5-super-vl \
  --tensor-parallel-size 4 \
  --trust-remote-code \
  --async-scheduling \
  --allowed-local-media-path=/ \
  --max-num-seqs 32 \
  --enable-mamba-shared-prefix-checkpoint \
  --limit-mm-per-prompt '{"video": 1, "image": 1}' \
  --video-pruning-rate 0.5 \
  --media-io-kwargs '{"video": {"fps": 2, "num_frames": 256}}' \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder \
  --reasoning-parser nemotron_v3 \
  --speculative_config.method mtp \
  --speculative_config.num_speculative_tokens 5

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#2%C3%97-gb200-up-to-1m-context-length2× GB200 (up to 1M Context Length)

Use this command if you need to serve the model with up to 1M context length.

vllm serve $MODEL \
  --served-model-name nemotron-3.5-super-vl \
  --tensor-parallel-size 2 \
  --trust-remote-code \
  --async-scheduling \
  --allowed-local-media-path=/ \
  --max-num-seqs 32 \
  --enable-mamba-shared-prefix-checkpoint \
  --limit-mm-per-prompt '{"video": 1, "image": 1}' \
  --video-pruning-rate 0.5 \
  --media-io-kwargs '{"video": {"fps": 2, "num_frames": 256}}' \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder \
  --reasoning-parser nemotron_v3 \
  --speculative_config.method mtp \
  --speculative_config.num_speculative_tokens 5

For the multimodal examples, each request can contain one image and one video. The video configuration samples up to 256 frames at 2 FPS with a pruning rate of 0.5. For 1080p videos, use 1 FPS and up to 128 frames; 2 FPS and 256 frames can be used for lower-resolution video such as 720p. Start with short MP4 clips of one to two minutes.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#sglangSGLang

Uselmsysorg/sglang:v0\.5\.20. The example below runs ontwo GB200 GPUs.

docker run --rm --gpus all --ipc=host \
  -p 8000:8000 \
  -e HF_TOKEN \
  -v "$HOME/.cache/huggingface:/root/.cache/huggingface" \
  --entrypoint sglang lmsysorg/sglang:v0.5.20 \
  serve --model-path "$MODEL_CKPT" \
  --served-model-name nemotron-3.5-super-vl \
  --host 0.0.0.0 \
  --port 8000 \
  --trust-remote-code \
  --tp-size 2 \
  --reasoning-parser nemotron_3 \
  --tool-call-parser qwen3_coder \
  --mamba-ssm-dtype float16 \
  --enable-mamba-cache-stochastic-rounding \
  --mamba-cache-philox-rounds 5 \
  --speculative-algorithm EAGLE \
  --speculative-num-steps 5 \
  --speculative-eagle-topk 1

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#api-clientAPI Client

Install the OpenAI-compatible client:

pip install openai

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#text-and-reasoningText and reasoning

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
model = "nemotron-3.5-super-vl"

response = client.chat.completions.create(
    model=model,
    messages=[{"role": "user", "content": "Explain why binary search takes logarithmic time."}],
    temperature=1.0,
    top_p=0.95,
    max_tokens=16384,
    extra_body={"chat_template_kwargs": {"enable_thinking": True}},
)
choice = response.choices[0]
if choice.finish_reason == "length":
    raise RuntimeError("The response reached max_tokens; increase the output budget.")
print(choice.message.content)

Setenable\_thinkingtoFalseto disable reasoning. For medium reasoning effort, use\{"enable\_thinking": True, "medium\_effort": True\}. Reasoning tokens and the final answer share themax\_tokensbudget.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#imagesImages

Use a multimodal server configuration. The example below sends a local image as a base64 data URL.

import base64
from pathlib import Path
from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
image_bytes = Path("image.png").read_bytes()
image_url = "data:image/png;base64," + base64.b64encode(image_bytes).decode()

response = client.chat.completions.create(
    model="nemotron-3.5-super-vl",
    messages=[{"role": "user", "content": [
        {"type": "text", "text": "Describe the image and any visible text."},
        {"type": "image_url", "image_url": {"url": image_url}},
    ]}],
    temperature=1.0,
    top_p=0.95,
    max_tokens=16384,
)
if response.choices[0].finish_reason == "length":
    raise RuntimeError("The response reached max_tokens; increase the output budget.")
print(response.choices[0].message.content)

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#video-with-vllmVideo with vLLM

Use a multimodal vLLM configuration with the video sampling settings above.

import base64
from pathlib import Path
from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
video_bytes = Path("video.mp4").read_bytes()
video_url = "data:video/mp4;base64," + base64.b64encode(video_bytes).decode()

response = client.chat.completions.create(
    model="nemotron-3.5-super-vl",
    messages=[{"role": "user", "content": [
        {"type": "text", "text": "Summarize the events in this video in order."},
        {"type": "video_url", "video_url": {"url": video_url}},
    ]}],
    temperature=1.0,
    top_p=0.95,
    max_tokens=16384,
)
if response.choices[0].finish_reason == "length":
    raise RuntimeError("The response reached max_tokens; increase the output budget.")
print(response.choices[0].message.content)

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#tool-callingTool calling

The model requests a tool call; the application executes it and returns the result. This complete example uses a local inventory lookup.

import json
from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
model = "nemotron-3.5-super-vl"
inventory = {"keyboard": 12, "monitor": 5}
tools = [{"type": "function", "function": {
    "name": "get_inventory",
    "description": "Return the inventory count for a product.",
    "parameters": {
        "type": "object",
        "properties": {"product": {"type": "string", "enum": ["keyboard", "monitor"]}},
        "required": ["product"],
        "additionalProperties": False,
    },
}}]
messages = [{"role": "user", "content": "How many keyboards are in stock?"}]

def complete(history, tool_choice="auto"):
    response = client.chat.completions.create(
        model=model,
        messages=history,
        tools=tools,
        tool_choice=tool_choice,
        temperature=1.0,
        top_p=0.95,
        max_tokens=16384,
        extra_body={"chat_template_kwargs": {"enable_thinking": True}},
    )
    choice = response.choices[0]
    if choice.finish_reason == "length":
        raise RuntimeError("The response was truncated; increase max_tokens.")
    return choice.message

message = complete(messages)
messages.append(message.model_dump(exclude_none=True))
if not message.tool_calls:
    print(message.content)
else:
    for call in message.tool_calls:
        arguments = json.loads(call.function.arguments)
        if (call.function.name != "get_inventory"
                or not isinstance(arguments, dict)
                or set(arguments) != {"product"}
                or not isinstance(arguments["product"], str)
                or arguments["product"] not in inventory):
            raise ValueError("Invalid inventory tool call")
        product = arguments["product"]
        messages.append({
            "role": "tool",
            "tool_call_id": call.id,
            "content": json.dumps({"product": product, "count": inventory[product]}),
        })
    print(complete(messages, tool_choice="none").content)

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#training-testing-and-evaluation-datasetsTraining, Testing, and Evaluation Datasets

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#trainingTraining

**Data Modality:**Text, images, and video **Foundation Training Data Size:More than 20 trillion tokensImage Training Data Size:1 Million to 1 Billion ImagesText Training Data Size:More than 10 Trillion TokensVideo Training Data Size:10,000 to 1 Million HoursFoundation Data Collection Period:**2013 to December 2025 **Data Collection Method:**Hybrid: automated, manually-collected, and synthetic **Labeling Method:**Hybrid: automated, manually-labeled, and synthetic

The foundation corpus contains curated and synthetic text covering English, 19 other spoken languages, and 43 programming languages. Sources include webpages, dialogue, articles, code, and other written materials across math, science, legal, finance, and general knowledge domains.

The post-training corpus includes reasoning, coding, tool use, instruction following, structured outputs, chat, safety, and long-context tasks, together with image and video understanding. Synthetic trajectories are filtered for structural correctness, repetition, and task validity.

For more information about the datasets used to train this model, please see thePublic Summary of Training Content

Detailed dataset information#### https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#base-pre-training-corpus-nemotron-3-foundationBase Pre-Training Corpus (Nemotron 3 Foundation)

The foundation of the model is trained on the Nemotron 3 corpus, comprising the following datasets from theNemotron Pretraining Datasets collection:

Dataset CollectionToken CountsDescriptionNemotron-CC-v2&v2.19.1TA massive collection of English web data filtered from Common Crawl, including 2.5T+ tokens of new organic, translated, and synthetically rephrased content.Nemotron-CC-Code-v1427.9BHigh-quality code tokens extracted from Common Crawl using the Lynx + LLM pipeline to preserve structure and equations.Nemotron-Pretraining-Code-v1&v2&v31.7TCurated GitHub code references with multi-stage filtering, deduplication, and large-scale synthetic code data.Nemotron-CC-Math-v1133.3BHigh-quality math pre-training dataset preserving LaTeX formatting and mathematical structures.Nemotron-Pretraining-Specialized-v1&v1.1&v1.2&Nemotron-Pretraining-SFT-v1660.0BSynthetic datasets targeting specialized domains such as STEM reasoning and scientific coding.Nemotron-Pretraining-Legal-v14.3BSynthetic datasets targeting the legal domain.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#public-datasetsPublic Datasets

DatasetCollection PeriodGSM8K4/23/2025CC-NEWS4/23/2025Common Crawl4/23/2025Wikimedia4/23/2025Bespoke-Stratos-17k4/23/2025tigerbot-kaggle-leetcodesolutions-en-2k4/23/2025glaive-function-calling-v24/23/2025APIGen Function-Calling4/23/2025LMSYS-Chat-1M4/23/2025Open Textbook Library - CC BY-SA & GNU subsetandOpenStax - CC BY-SA subset4/23/2025Advanced Reasoning Benchmark,tigerbot-kaggle-leetcodesolutions-en-2k,PRM800K, andSciBench4/23/2025FineWeb-24/23/2025Court ListenerLegacy DownloadpeS2oLegacy DownloadOpenWebMathLegacy DownloadBioRxivLegacy DownloadPMC Open Access SubsetLegacy DownloadOpenWebText2Legacy DownloadStack Exchange Data DumpLegacy DownloadPubMed AbstractsLegacy DownloadNIH ExPorterLegacy DownloadarXivLegacy DownloadBigScience Workshop DatasetsLegacy DownloadSEC’s Electronic Data Gathering, Analysis, and Retrieval (EDGAR)Legacy DownloadAdvanced Mathematical Problem SolvingLegacy DownloadMathPileLegacy DownloadNuminaMath CoTLegacy DownloadPMC ArticleLegacy DownloadFLANLegacy DownloadAdvanced Reasoning BenchmarkLegacy DownloadSciBenchLegacy DownloadWikiTableQuestionsLegacy DownloadFinQALegacy DownloadRiddlesLegacy DownloadProblems in Elementary Mathematics for Home StudyLegacy DownloadMedMCQALegacy DownloadCosmos QALegacy DownloadMCTestLegacy DownloadAI2’s Reasoning ChallengeLegacy DownloadOpenBookQALegacy DownloadMMLU Auxiliary TrainLegacy Downloadsocial-chemestry-101Legacy DownloadMoral StoriesLegacy DownloadThe Common Pile v0.1Legacy DownloadFineMathLegacy DownloadMegaMathLegacy DownloadMultiverseMathHard10/2/2025SWE-Gym10/2/2025WorkBench10/2/2025WildChat-1M10/2/2025OpenCodeReasoning-210/2/2025HelpSteer310/2/2025opc-sft-stage210/2/2025Big-Math-RL-Verified10/2/2025MetaMathQA10/2/2025simple-arithmetic-problems10/2/2025arithmetic10/2/2025Skywork-OR1-RL-Data10/2/2025FastChat10/2/2025News Commentary10/2/2025Essential-Web10/2/2025finepdfs10/2/2025HotpotQA10/2/2025SQuAD2.010/2/2025NLTK Words Lists10/2/2025NVClip (NV-CC-Img-Text-Dataset)2022-2026ScaleCUA, Processed Training Subset2022-2026RoboInter-VQA2022-2026Open Images V6, Object-Detection Training Records2022-2026YT1B PLM Video QA, Custom No-Thinking and Safety-Filtered2022-2026YT1B PLM Dense Captions, No-Thinking and Safety-Filtered2022-2026RoboVQA2022-2026YT1B PLM Thinking Traces, Safety-Filtered2022-2026Boyi Short-Video QA2022-2026Boyi Short-Video QA, Custom No-Thinking and Safety-Filtered2022-2026K710 Video Classification, Full Training Snapshot2022-2026Molmo2-VideoCapQA2022-2026Molmo2-VideoCapQA, Custom No-Thinking and Safety-Filtered2022-2026Molmo2-VideoPoint, Custom No-Thinking and Safety-Filtered2022-2026Molmo2-VideoPoint2022-2026

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#crawled-and-scraped-from-online-sources-by-nvidiaCrawled and Scraped from Online Sources by NVIDIA

The English Common Crawl data was downloaded from the Common Crawl Foundation (see their FAQ for details on their crawling) and includes the snapshots CC-MAIN-2013-20 through CC-MAIN-2025-13. The data was subsequently deduplicated and filtered in various ways used in the Nemotron-CC data pipeline. Additionally, we extracted data for fifteen languages from the following three Common Crawl snapshots: CC-MAIN-2024-51, CC-MAIN-2025-08, CC-MAIN-2025-18. The fifteen languages included were Arabic, Chinese, Danish, Dutch, French, German, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish, Swedish, and Thai. As we did not have reliable multilingual model-based quality classifiers available, we applied just heuristic filtering instead—similar to what we did for lower quality English data in the Nemotron-CC pipeline, but selectively removing some filters for some languages that did not work well. Deduplication was done in the same way as for Nemotron-CC.

The GitHub Crawl was collected using the GitHub REST API and the Amazon S3 API. Each crawl was operated in accordance with the rate limits set by its respective source, either GitHub or S3. We collect raw source code and subsequently remove any having a license which does not exist in our permissive-license set.

DatasetModalityDataset SizeCollection PeriodCollecting OrganisationEnglish Common CrawlText3.36T4/8/2025NVIDIA Advanced Deep Learning ResearchEnglish Common Crawl 1.1TextNot disclosed10/2/2025NVIDIA Advanced Deep Learning ResearchMultilingual Common CrawlText812.7B5/1/2025NVIDIA Advanced Deep Learning ResearchGitHub CrawlText747.4B4/29/2025NVIDIA Advanced Deep Learning ResearchGitHub Crawl 1.1Text172.7B9/30/2025NVIDIA Advanced Deep Learning ResearchSangrahaText109,263,115 indexed source sequences: 77,231,480 verified and 32,031,635 unverifiedUndisclosedNVIDIA Advanced Deep Learning ResearchLITA Chunked Video Labels, Custom No-Thinking Dense CaptionsText (video labels)151,349 recordsUndisclosedNVIDIA Advanced Deep Learning ResearchLITA Chunked Video Labels, Cleaned v2Text (video labels)150,434 recordsUndisclosedNVIDIA Advanced Deep Learning ResearchSingle-Page Document Labels, Qwen3.5 v2Text and images50,930 recordsUndisclosedNVIDIAProcessed YouTube Videos, Cosmos Dense CaptionsText and video274,472 recordsUndisclosedNVIDIA

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#private-non-publicly-accessible-datasets-of-third-partiesPrivate Non-publicly Accessible Datasets of Third Parties

DatasetModel(s) usedGlobal RegulationUnknownTAUS Translation MemoryUnknownScale HLEUnknownHackerRank CodingUnknownRL data for SearchGemini 3; GPT-5Mercor SWE-AgentsV1UndisclosedKeenSight Batch 1, Relabeled SubsetQwen3.5-122B-A10BKeenSight Batch 1, Remaining SubsetQwen3.5-122B-A10BCamio Multilingual OCR and Document LayoutUndisclosedChinese Visual Question Answering, Relabeled SubsetQwen3.5-122B-A10BCamio Chinese OCR, Cleaned VariantQwen3.5-122B-A10BCamio English OCR, Cleaned VariantQwen3.5-122B-A10BKeenSight Batch 2, Remaining SubsetQwen3.5-122B-A10BKeenSight Batch 1, Safety-Filtered Trace VariantQwen3.5-122B-A10BChinese Visual Question Answering, Safety-Filtered Trace VariantQwen3.5-122B-A10BCamio Chinese Question Answering, Verified-Correct SubsetQwen3.5KeenSight Chinese OCR, Filtered SubsetDeepSeek-OCRCamio Synthetic Trading-Graph Question AnsweringUndisclosedCamio Synthetic Menu and Receipt Question AnsweringQwen3.5-122B-A10BCamio Chinese OCR, Formatted VariantQwen3.5-122B-A10BCamio Synthetic Time-Series Question AnsweringUndisclosedCamio English OCR, Formatted VariantQwen3.5-122B-A10BLean Formal Proofs — RLUndisclosedDocument-Grounded Finance Reasoning, Patch Tasks — RLUndisclosedMath CoT — SFTKimi K3Chat — SFTUndisclosedChat, Reasoning Off — SFTUndisclosedConversational Tool-Use Simulation — RLUndisclosedMath CoT, Low Variant — SFTKimi K3Document-Grounded Finance Reasoning — SFTUndisclosedSafety Preference v0.3 — RLUndisclosedResponse-Length Instruction Following, Reasoning Off — SFTUndisclosedMath Python TIR, Low Variant — SFTKimi K3Math Python TIR — SFTKimi K3General-Purpose Tool Calling — SFTQwen3.5-397B-A17B; Gemma-4-31B-itKnowledge Agent — SFTUndisclosedDataFunnel arXiv Science4 v1.0 — RLDeepSeek-V4-Flash (problem extraction); GLM-5.2 (refinement); Kimi K3 and K3 V5 (verification and filtering); Super 3.5 r14 (difficulty profiling); Qwen3-235B-A22B-Instruct-2507-FP8 (reward judging)Structured Outputs v3 — RLgpt-oss-120bResponse-Length Instruction Following — SFTUndisclosedFiltered CUDA Coding — SFTUndisclosedReasoning Gym — RLUndisclosedCompetitive Coding Challenges v2 — SFTGLM-5.2Multi-Turn Instruction Following, Reasoning Off — SFTUndisclosedMulti-Turn Instruction Following — SFTUndisclosedInstruction Following, Filtered RLHF Prompts — SFTUndisclosedCitation Formatting — RLgpt-oss-120bLong-Context General Data — SFTUndisclosedFree-Form Formatting — RLgpt-oss-120bStructured Outputs v4 — RLUndisclosedTool-Call Schema — RLNo model used for RL conversion; source-generation models undisclosedCompetitive Coding, OCR Python — SFTDeepSeek-V4-ProLean Formal Proofs — SFTgpt-oss-120b; Goedel-Prover-V2-32B; Qwen2.5-32B-InstructCitation Instruction Following — RLUndisclosedSearch Graph Walks — SFTMiniMax-M2General-Purpose Tool Calling, Reasoning Off — SFTUndisclosedKnowledge Agent Tau3, Reasoning Off — SFTUndisclosedARC-AGI Puzzle Reasoning — SFTKimi-K2.5; Qwen3-235B-A22B-Thinking-2507; DeepSeek-V3.2; Qwen3-Next-80B-A3B-Thinking; GLM-4.5-Air-FP8; gpt-oss-120b; GLM-4.7-FP8; Ring-1T; Ministral-3-14B-Reasoning-2512Calendar — RLUndisclosedACEProof — SFTUndisclosedLong Translation — SFTUndisclosedAgentless Software Engineering — SFTUndisclosedLong-Context PDF Translation v1.0 — SFTDeepSeek-V4-Flash (translation generation); GPT-5.6 Sol (filtering and judging)KeenSight Batch 2, Relabeled and Filtered SubsetQwen3.5-122B-A10B; Gemini 3 Flash Preview (filtering)Camio Chinese Letter QA, Cleanedentrace\_cleanedVariantQwen3-VL-30B-A3B-Thinking-FP8; gpt-oss-120b; Qwen3.5-122B-A10B; GPT-4o mini (filtering and quality checks)Camio Chinese Letter QA, Cleaned VariantQwen3-VL-30B-A3B-Thinking-FP8; gpt-oss-120b; Qwen3.5-122B-A10B; GPT-4o mini (filtering and quality checks)Camio Chinese Text QA, Cleanedentrace\_cleanedVariantQwen3-VL-30B-A3B-Thinking-FP8; gpt-oss-120b; Qwen3.5-122B-A10B; GPT-4o mini (filtering and quality checks)

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#private-non-publicly-accessible-datasets-by-nvidiaPrivate Non-publicly Accessible Datasets by NVIDIA

DatasetModel(s) usedSimple MinesweeperUndisclosedSimple SudokuUndisclosedMultitool Typewriter HardUndisclosedMachine Translation of News Commentary and TAUS Translation MemoryUndisclosedMachine Translation of STEM -Qwen2.5-14B-InstructCompetitive Coding RL data from Nemotron CascadeUndisclosedLong context RLUndisclosedSingle-step SWE RL for patch generationUndisclosedOpenHands SWEUndisclosed

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#nvidia-sourced-synthetic-datasets-pre-trainingNVIDIA-Sourced Synthetic Datasets (Pre-Training)

DatasetModalityDataset SizeSeed DatasetModel(s) used for generationNemotron-Pretraining-Fact-SeekingText35.0BFineWikiQwen3-30B-A3B-Instruct-2507Nemotron-Pretraining-LegalText4.3BCommonPile (caselaw_access_project_filtered); California Code of Regulations; Judicial Ethics Opinions;GLOBALCIT;CUAD; Nemotron Personas;ToSDR Terms of Service Corpus; CodeHima/TOS_Dataset;ContractNLI; CaseHOLD; Code of Federal Regulations;Canadian Case Law(subsets that allow commercial use)Qwen3-235B-A22B-Thinking-2507Nemotron-Pretraining-Formal-LogicText128MNemotron PersonasQwen3-235B-A22B-Thinking-2507Nemotron-Pretraining-EconomicsText73.4M-Qwen3-235B-A22B-Thinking-2507Nemotron-Pretraining-Multiple-ChoiceText1.6BMMLU Auxiliary TrainDeepSeek-V3;Qwen3-235B-A22BNemotron-Pretraining-Code-ConceptsText7.3B-gpt-oss-20b;gpt-oss-120bNemotron-Pretraining-Unconditional-AlgorithmicText196.5M-gpt-oss-120b;Qwen3-235B-A22BMore Synthetic Tasks from DeepSeek-V3 and Qwen3-235B-A22BText1.1Btrain splits of acp_bench; ai2_arc; babi; gsm8k; hendrycks_math; IFEval; MedText; mediqa_qa; mlqa; MMLU-Pro; mmlu-pro-plus; MMLU-ProX; nq_open; tinyGSM8k; truthful_qa; truthfulqa-multi; MATH-lighteval; mmlu; awesome-chatgpt-prompts; super_glueDeepSeek v3;Qwen3-235B-A22BSynthetic Tasks from DeepSeek-V3 and Qwen3-235B-A22BText6.7Btrain splits of Into the Unknown; AI2 ARC (AI2 Reasoning Challenge); BLiMP (Benchmark of Linguistic Minimal Pairs); CommonSenseQA; GLUE; HeadQA; Hendrycks Ethics; Memo Trap; modus-tollens; NeQA; pattern-matching-suppression; mastermind_24_mcq_random; mastermind_24_mcq_close; quote-repetition; redefine-math; Repetitive Algebra; sig-figs; MMLU-Pro; MC-TACO; MedConceptsQA; MMLU_dataset; OpenbooksQA; PIQA (Physical Interaction Question Answering); SocialIQA; SuperGLUE; tinyAI2_arc; tinyMMLU; tinyWinogrande; TruthfulQA; WebQuestions; Winogrande; GPQA; MBPPDeepSeek v3;Qwen3-235B-A22BSynthetic Art of Problem Solving from DeepSeek-R1Text40BArt of Problem Solving;American Mathematics Competitions 8;American Mathematics Competitions 10DeepSeek-R1Synthetic Moral Stories and Social Chemistry from Qwen3-235B-A22B-Thinking-2507 and Mixtral-8x22B-v0.1Text15.2Msocial-chemestry-101;Moral StoriesQwen3-235B-A22B-Thinking-2507;Mixtral-8x22B-v0.1Synthetic Moral Stories and Social Chemistry from Mixtral-8x22B-v0.1Text327Msocial-chemestry-101;Moral StoriesMixtral-8x22B-v0.1Synthetic Social Sciences seeded with OpenStax from DeepSeek-V3, Mixtral-8x22B-v0.1, and Qwen2.5-72BText83.6MOpenStax - CC BY-SA subsetDeepSeek-V3;Mixtral-8x22B-v0.1;Qwen2.5-72BSynthetic Health Sciences seeded with OpenStax from DeepSeek-V3, Mixtral-8x22B-v0.1, and Qwen2.5-72BText9.7MOpenStax - CC BY-SA subsetDeepSeek-V3;Mixtral-8x22B-v0.1;Qwen2.5-72BSynthetic STEM seeded with OpenStax, Open Textbook Library, and GSM8K from DeepSeek-R1, DeepSeek-V3, DeepSeek-V3-0324, and Qwen2.5-72BText175MOpenStax - CC BY-SA subset;GSM8K;Open Textbook Library - CC BY-SA & GNU subsetDeepSeek-R1,DeepSeek-V3;DeepSeek-V3-0324;Qwen2.5-72BNemotron-PrismMathText4.6BBig-Math-RL-Verified;OpenR1-Math-220kQwen2.5-0.5B-instruct,Qwen2.5-72B-Instruct;DeepSeek-R1-Distill-Qwen-32BSynthetic Question Answering Data from Papers and Permissible Books from Qwen2.5-72B-InstructText350MarXiv;National Institutes of Health ExPorter;BioRxiv;PMC Article;USPTO Backgrounds;peS2o; Global Regulation;CORE;PG-19;DOAB CC BY & CC BY-SA subset;NDLTDQwen2.5-72B-InstructSynthetic RephrasedMath Data from Common Crawlfrom phi-4Text73BCommon Crawlphi-4Synthetic Math Data from Common Crawl 4plusText52.3BCommon Crawlphi-4Synthetic Math Data from Common Crawl 3Text80.9BCommon Crawlphi-4Synthetic AGIEval seeded with AQUA-RAT, LogiQA, and AR-LSAT from DeepSeek-V3 and DeepSeek-V3-0324Text4.0BAQUA-RAT;LogiQA;AR-LSATDeepSeek-V3;DeepSeek-V3-0324Synthetic AGIEval seeded with AQUA-RAT, LogiQA, and AR-LSAT from Qwen3-30B-A3BText4.2BAQUA-RAT;LogiQA;AR-LSATQwen3-30B-A3BSynthetic Art of Problem Solving from Qwen2.5-32B-Instruct, Qwen2.5-Math-72B, Qwen2.5-Math-7B, and Qwen2.5-72B-InstructTextUndisclosedArt of Problem Solving;American Mathematics Competitions 8;American Mathematics Competitions 10;GSM8K;PRM800KQwen2.5-32B-Instruct;Qwen2.5-Math-72B;Qwen2.5-Math-7B;Qwen2.5-72B-InstructSynthetic MMLU Auxiliary Train from DeepSeek-R1Text0.5BMMLU Auxiliary TrainDeepSeek-R1Synthetic Long Context Continued Post-Training Data from Papers and Permissible Books from Qwen2.5-72B-InstructTextUndisclosedarXiv;National Institutes of Health ExPorter;BioRxiv;PMC Article;USPTO Backgrounds;peS2o; Global Regulation;CORE;PG-19;DOAB CC BY & CC BY-SA subset;NDLTDQwen2.5-72B-InstructSynthetic Common Crawl from Qwen3-30B-A3B and Mistral-Nemo-12B-InstructText415.8BCommon CrawlQwen3-30B-A3B;Mistral-NeMo-12B-InstructSynthetic Multilingual Data from Common Crawl from Qwen3-30B-A3BTextUndisclosedCommon CrawlQwen3-30B-A3BSynthetic Multilingual Data from Wikimedia from Qwen3-30B-A3BTextUndisclosedWikimediaQwen3-30B-A3BSynthetic Math Data from Wikimedia from Nemotron-4-340B-InstructTextUndisclosed-Nemotron-4-340B-InstructSynthetic Common Crawl Code from phi-4Text427.9BCommon Crawlphi-4Synthetic Scientific Coding from Qwen3-235B-A22BText1.2BWikimediaQwen3-235B-A22BTool Calling DataText26.2B-Qwen3-235B-A22B-2507;gpt-oss-120bSynthetic Essential-Web from QwQ-32BText28.1BEssential-WebQwQ-32BTranslated Synthetic CrawlText389.9BCommon CrawlQwen3-30B-A3BTranslated Synthetic WikipediaText7.9BWikimediaQwen3-30B-A3BSynthetic Long Context from Qwen3-235B-A22B-Instruct-2507TextUndisclosedCORE;PG-19;DOAB CC BY & CC BY-SA subset;NDLTDQwen3-235B-A22B-Instruct-2507Synthetic Search STEM OPENQ from DeepSeek-R1-0528TextUndisclosed-DeepSeek-R1-0528Synthetic MCQ from Qwen2.5-32B-Instruct and DeepSeek-R1-0528TextUndisclosed-Qwen2.5-32B-Instruct;DeepSeek-R1-0528Synthetic Offline Search MCQA HLE from DeepSeek-R1-0528TextUndisclosed-DeepSeek-R1-0528Synthetic Offline Search MCQA GPQA from Qwen3-235B-A22B and DeepSeek-R1-0528TextUndisclosed-Qwen3-235B-A22B;DeepSeek-R1-0528Synthetic Human Preference from QwQ-32B, Qwen3-30B-A3B, Qwen3-235B-A22B, Qwen3-235B-A22B-Instruct-2507, Mistral-Small-3.1-24B-Instruct-2503, Mistral-Small-3.2-24B-Instruct-2506, MiniMax-M1-80k, MiniMax-M1-40k, Kimi-K2-Instruct, DeepSeek-V3-0324, DeepSeek-R1-0528TextUndisclosed-QwQ-32B;Qwen3-30B-A3B;Qwen3-235B-A22B;Qwen3-235B-A22B-Instruct-2507;Mistral-Small-3.1-24B-Instruct-2503;Mistral-Small-3.2-24B-Instruct-2506;MiniMax-M1-80k;MiniMax-M1-40k;Kimi-K2-Instruct;DeepSeek-V3-0324;DeepSeek-R1-0528Synthetic WildChat-1M and arena-human-preference-140k from DeepSeek-R1, gemma-2-2b-it, gemma-3-27b-it, gpt-oss-20b, gpt-oss-120b, Mistral-7B-Instruct-v0.3, Mixtral-8x22B-Instruct-v0.1, Nemotron-4-340B-Instruct, NVIDIA-Nemotron-Nano-9B-v2, Phi-4-mini-instruct, Phi-3-small-8k-instruct, Phi-3-medium-4k-instruct, Qwen3-235B-A22B, QwQ-32BTextUndisclosedWildChat-1M;arena-human-preference-140kDeepSeek-R1;gemma-2-2b-it;gemma-3-27b-it;gpt-oss-20b;gpt-oss-120b;Mistral-7B-Instruct-v0.3;Mixtral-8x22B-Instruct-v0.1;Nemotron-4-340B-Instruct;NVIDIA-Nemotron-Nano-9B-v2;Phi-4-mini-instruct;Phi-3-small-8k-instruct;Phi-3-medium-4k-instruct;Qwen3-235B-A22B;QwQ-32BSynthetic Code from Qwen3-32BTextUndisclosedEnglish Common Crawl; English Common Crawl 1.1Qwen3-32BSynthetic OpenCodeReasoning from DeepSeek-R1TextUndisclosedOpenCodeReasoningDeepSeek-R1Synthetic OpenCodeReasoning from DeepSeek-R1-0528TextUndisclosedOpenCodeReasoningDeepSeek-R1-0528Synthetic HackerRank Coding from DeepSeek-R1-0528TextUndisclosedHackerRank Coding DatasetDeepSeek-R1-0528Synthetic LIMO from DeepSeek-R1-0528TextUndisclosedLIMODeepSeek-R1-0528Synthetic SCP from DeepSeek-R1-0528TextUndisclosedSCP-116KDeepSeek-R1-0528Synthetic Stack Exchange from DeepSeek-R1-0528TextUndisclosedStack ExchangeDeepSeek-R1-0528Synthetic Stack Exchange from gpt-oss-120b and Qwen2.5-32B-InstructTextUndisclosedStack Exchangegpt-oss-120b;Qwen2.5-32B-InstructSynthetic Stack Exchange from gpt-oss-120bTextUndisclosedStack Exchangegpt-oss-120bSynthetic Art of Problem Solving from gpt-oss-120b and Qwen2.5-32B-InstructTextUndisclosedArt of Problem Solving;American Mathematics Competitions 8;American Mathematics Competitions 10gpt-oss-120b;Qwen2.5-32B-InstructSynthetic Common Crawl from Qwen3-30B-A3BTextUndisclosedCommon CrawlQwen3-30B-A3BSynthetic Wikipedia from Qwen3-30B-A3BTextUndisclosedWikimediaQwen3-30B-A3BSynthetic Essential-Web from Qwen3-30B-A3B and Qwen3-235B-A22B-Thinking-2507TextUndisclosedEssential-WebQwen3-30B-A3B;Qwen3-235B-A22B-Thinking-2507Synthetic Essential-Web from gpt-oss-120bTextUndisclosedEssential-Webgpt-oss-120bSynthetic Textbook Math from Qwen3-30B-A3B, Qwen3-235B-A22B, phi-4TextUndisclosedCommon Crawl;FineMathQwen3-30B-A3B;Qwen3-235B-A22B;phi-4Synthetic Math and Code from DeepSeek-R1 and DeepSeek-R1-0528TextUndisclosedMagicoder-Evol-Instruct-110K;opc-sft-stage2;TACO;OpenCodeReasoning;OpenMathReasoning;NuminaMath CoTDeepSeek-R1;DeepSeek-R1-0528Synthetic Math from gpt-oss-120b and Qwen2.5-32B-InstructTextUndisclosed-gpt-oss-120b;Qwen2.5-32B-InstructSynthetic OpenMathReasoning from gpt-oss-120b and Qwen2.5-32B-InstructTextUndisclosedOpenMathReasoninggpt-oss-120b;Qwen2.5-32B-InstructSynthetic KernelBook from DeepSeek-R1-0528TextUndisclosedKernelBookDeepSeek-R1-0528Synthetic Scale HLE from gpt-oss-120bTextUndisclosedScale HLEgpt-oss-120bSynthetic CDQuestions from gpt-oss-120bTextUndisclosedCDQuestionsgpt-oss-120bSynthetic GPQA from gpt-oss-120b and Qwen2.5-32B-InstructTextUndisclosedStack Exchangegpt-oss-120b;Qwen2.5-32B-InstructSynthetic Vedantu from gpt-oss-120bTextUndisclosedVedantugpt-oss-120bSynthetic Search STEM MCQ from Qwen3-235B-A22B and DeepSeek-R1-0528TextUndisclosed-Qwen3-235B-A22B;DeepSeek-R1-0528Synthetic OpenSTEM from Qwen2.5-32B-Instruct and DeepSeek-R1-0528TextUndisclosed-Qwen2.5-32B-Instruct;DeepSeek-R1-0528Synthetic MCQ10 from DeepSeek-R1-0528TextUndisclosed-DeepSeek-R1-0528Synthetic MCQ4 from Qwen3-235B-A22B, DeepSeek-R1-0528, and Qwen3-235B-A22B-Instruct-2507TextUndisclosed-Qwen3-235B-A22B;DeepSeek-R1-0528;Qwen3-235B-A22B-Instruct-2507Indic SDG for NemotronText711,276,565 indexed source-sequence entries; final training-blend count not confirmednvidia/Nemotron-CC-v2google/gemma-4-31B-it; google/gemma-4-31B-it-assistant

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#nvidia-sourced-synthetic-datasets-post-trainingNVIDIA-Sourced Synthetic Datasets (Post-Training)

DatasetModalityDataset SizeSeed DatasetModel(s) used for generationSynthetic Single-Step Tool Use — RLTextApproximately 1,433,203 samplesHandwritten prompts; no external seed dataDeepSeek-V3.2; DeepSeek-R1-0528; Qwen3-235B-A22B-Thinking-2507; Qwen3-32BSynthetic Sparse Tables with Mixed PromptsText and images100,000 recordsGenerated in-house; no external seed dataNone; generated without an SDG modelSynthetic Single-Step Tool Use — RLTextApproximately 1,433,203 samplesHandwritten prompts; no external seed dataDeepSeek-V3.2; DeepSeek-R1-0528; Qwen3-235B-A22B-Thinking-2507; Qwen3-32BSynthetic Sparse Tables with Mixed PromptsText and images100,000 recordsGenerated in-house; no external seed dataNone; generated without an SDG modelSynthetic Competitive MATH Proofs from DeepSeek-V4-ProTextUndisclosed[AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions][deepseek-ai/DeepSeek-V4-Pro]Synthetic Hermes Agent Reasoning TracesTextUndisclosed[lambda/hermes-agent-reasoning-traces][hermes-agent-generator]Synthetic Competitive Coding from DeepSeek-V4-ProTextUndisclosed[NVCompetitiveCodingV1][deepseek-ai/DeepSeek-V4-Pro]Synthetic Competitive Science Reasoning from DeepSeek-V4-ProTextUndisclosed[AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [EssentialAI/essential-web-v1.0]; [cdquestions.com]; [Pile-FreeLaw]; [Vedantu]; [askfilo]; [doubtnut]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)]; [AAPT]; [ChemData 700K]; [oMeBench]; [Flavor Analysis and Recognition Transformer]; [ChemCoTBench]; [Llama Nemotron Dataset][deepseek-ai/DeepSeek-V4-Pro]Synthetic Competitive MATH CoT and TIR from Nemotron 5.5TextUndisclosed[Pile-FreeLaw]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions][Nemotron 5.5]Vendor Terminal Bench-like Tasks from MercorTextUndisclosed[Terminal bench like tasks curated by the vendor][Undisclosed - purchased dataset]Turing Math Data PackTextUndisclosed[Turing Math Data Pack dataset][Undisclosed - purchased dataset]Synthetic Holdout, Skywork, DAPO, and Turing Math from GPT-5.5TextUndisclosed[DocQA-RL-1.6K]; [DAPO-Math-17k][GPT-5.5]Synthetic Long Context RL from QwenLong L1 and DocQA-RL-1.6KTextUndisclosed[DocQA-RL-1.6K]UndisclosedSynthetic Competitive Coding Gym TasksTextUndisclosed[NVCompetitiveCodingV1.1]UndisclosedSynthetic Finance SEC Search Agent from GPT-OSS-120B and Qwen3TextUndisclosed[SEC filings from sec.gov][GPT-OSS-120B]; [Qwen3-235B-A22B-Instruct]; [Qwen3-4B-Instruct]Synthetic Structured Outputs from Qwen3-30B-A3B-Instruct-2507, Qwen3-30B-A3B-Thinking-2507, Qwen3-235B-A22B-Instruct-2507, and Qwen3-235B-A22B-Thinking-2507TextUndisclosed[Nemotron-RL-agent-structured-outputs-v1][Qwen3-30B-A3B-Instruct-2507]; [Qwen3-235B-A22B-Instruct-2507]Synthetic Long Context Equivalence Rule from Qwen3-235B-A22B-Thinking-2507 and DeepSeek-R1TextUndisclosed[Long-context SFT data][Qwen/Qwen3-235B-A22B-Thinking-2507]; [Deepseek-ai/DeepSeek-R1]Synthetic Science RL Data Blend from Qwen2.5-32BTextUndisclosed[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][Qwen2.5-32B]Synthetic Abstention Data from Nemotron Super v3TextUndisclosed[Go abstention Dataset][nvidia/nvidia/nemotron-3-super-v3]Synthetic Chemistry Data from Nemotron Super v3TextUndisclosed[ChemData 700K][nvidia/nvidia/nemotron-3-super-v3]Synthetic Tool Call Schema for RLText469,983[UltraTool]; [ToolEyes]; [AutoTools]; [API-Bank]; [Nemotron-Personas-USA]; [Salesforce xLAM function-calling]; [Glaive function-calling-v2]; [Agent-Ark/Toucan-1.5M][DeepSeek-V3.2]; [GLM-4.6]; [gpt-oss-120b]; [Kimi-K2-Instruct]Synthetic Freeform Text Formatting from GPT-OSS-120BTextUndisclosed[In-house data][GPT OSS 120B - Apache 2.0]Synthetic Citation Formatting from GPT-OSS-120BTextUndisclosed[In-house data][GPT OSS 120B - Apache 2.0]Droid Harness Pivot Vendor DataTextUndisclosed[Droid Harness Pivot vendor data]UndisclosedSynthetic HotpotQA Training Data from Qwen3-235BTextUndisclosed[HotpotQA][Qwen3-235B]Synthetic Natural Language Math Proofs from Nemotron 5.5TextUndisclosed[AMC8, AMC10, and AIME problem sets hosted on Art of Problem Solving]; [Pile-StackExchange][Nemotron 5.5]Synthetic Stack Overflow OpenQTextUndisclosed[Pile-FreeLaw]UndisclosedChemistry Ether0 Vendor DataTextUndisclosed[Chemistry ether0 vendor data]UndisclosedSynthetic Litmus-Bench Chemistry from ChEMBLTextUndisclosed[ChEMBL]; [Nemo Gym RL dataset generated from ChEMBL with RDKit]UndisclosedSynthetic ZINC Chemistry from Nemotron Super v3TextUndisclosed[ZINC][Nemotron Super v3]ARC-AGI Gym EnvironmentTextUndisclosed[ARC-AGI-2][ARC-AGI-2]Synthetic Agentic Search Tool-Use from DeepSeek-V3.2TextUndisclosed[Mercor Data][DeepSeek-V3.2]Synthetic Text-To-SQLText96,564[In-house Text-to-SQL data][gpt-oss-120b]Dialog Memory Vendor DataTextUndisclosed[Patronus external vendor agreement]UndisclosedSynthetic Indirect Prompt Injection from Nemotron Super v3 and Qwen3-Next-80B-A3B-InstructTextUndisclosed[In-house indirect prompt injection data][nvidia/nemotron-3-super-v3, qwen/qwen3-next-80b-a3b-instruct]Synthetic Malicious Code and Agentic SecurityTextUndisclosed[In-house malicious-code / agentic-security data]UndisclosedSynthetic Single-Step SWE Patch SelectionTextUndisclosed[SWE-Gym Dataset]; [SWE Bench Verified Benchmark][ground truth and task checks]Synthetic Natural Language Math Final Answers from Nemotron 5.5TextUndisclosed[AMC8, AMC10, and AIME problem sets hosted on Art of Problem Solving]; [Pile-StackExchange][nemotron 5.5]Synthetic Simple Math Prompts for Token EfficiencyTextUndisclosed[In-house simple math prompts]UndisclosedSynthetic Abstention Data from Nemotron Super v3 (CRAG)TextUndisclosed[CRAG][nvidia/nvidia/nemotron-3-super-v3]Synthetic Agentless SWEText242,536[SWE-Rebench-V2]; [SWEbench Training Set]; [R2E-Gym/R2E-Gym-Subset]; [SWE-Gym/SWE-Gym]; [SWE-Rebench][openai/gpt-oss-120b]Synthetic Agentless SWE from DeepSeek-R1-0528Text209,976[SWE-Bench-Train]; [SWE-Fixer-Train]; [SWE-reBench]; [SWE-Smith][deepseek-ai/DeepSeek-R1-0528]Synthetic Agentic CUDA Traces from GLM-4.7Text2,276[Internal CUDA task data][GLM-4.7]Synthetic Math Proofs from DeepSeek-V3.2-SpecialeText820,772[Nemotron-Math-Proofs-v1][SDG: DeepSeek-V3.2-Speciale]; [Filter: proof validation]Synthetic Multilingual SFT from DeepSeek-V3Text1,245,284[Nano v3 SFT data][DeepSeek-V3]Synthetic Agentic Code from gpt-oss-120bText109,086[NVAgenticCLIPrompts-v1]; [NVAgenticSkills-v1]; [NVAgenticCLIMultiTurnPrompts-v1][openai/gpt-oss-120b]Synthetic Agentic CLI and Web Skills from gpt-oss-120bText27,418[NVAgenticCLIPrompts-v1]; [NVAgenticSkills-v1]; [NVAgenticCLIMultiTurnPrompts-v1]; [NVAgenticCLIPrompts-Web-v1][openai/gpt-oss-120b]Synthetic Agentic Coding from gpt-oss-120bText160,531[NVAgenticCLIPrompts-v1]; [NVAgenticSkills-v1]; [NVAgenticCLIMultiTurnPrompts-v1]; [NVAgenticCLIPrompts-Web-v1][openai/gpt-oss-120b]Synthetic OpenCode Agentic Tasks from gpt-oss-120bText614,773[NVAgenticCLIPrompts-v1]; [NVAgenticSkills-v1]; [NVAgenticCLIMultiTurnPrompts-v1]; [NVAgenticCLIPrompts-Web-v1][openai/gpt-oss-120b]Synthetic SWE UnverifiedTextUndisclosed[NVAgenticCLIPrompts-v1]; [NVAgenticSkills-v1]; [NVAgenticCLIMultiTurnPrompts-v1]; [NVAgenticCLIPrompts-Web-v1][gpt-oss-120b]; [Qwen/Qwen3-Coder-480B-A35B-Instruct]; [GLM-4.7-Flash]Synthetic ARC-AGI Ultra DataText192,016[ARC-AGI-2]; [arc dataset collection][ARC-AGI-2]Synthetic LiveCodeBench TIR from DeepSeek-R1-0528Text1,283,398[Nemotron-X training datasets][DeepSeek-R1-0528]Synthetic Verilog and SystemVerilog Code from DeepSeek-R1-0528 and GPT-OSS-120BText1,233,247[Verilog/SystemVerilog seed code][SDR: DeepSeek R1 0528 and GPT-OSS-120B]; [Filtering: Claude 4 Sonnet]Synthetic Aider Python Tasks from DeepSeek-R1-0528Text236,099[Exercism (GitHub Python)][Deepseek R1 0528]Synthetic Chat Reasoning-Off Data from GLM-5Text646,738[lmarena-ai/repochat-arena-preference-4k user prompts][Multi-turn conversations generated by GLM-5 with best-of-4 selection via Qwen3-Nemotron-235B-A22B-GenRM]Synthetic Chat Reasoning-On Data from GLM-5Text644,286[lmarena-ai/repochat-arena-preference-4k user prompts]; [lmarena-ai/arena-expert-5k user prompts]; [lmarena-ai/arena-human-preference-55k user prompts]; [lmarena-ai/arena-human-preference-100k user prompts]; [lmarena-ai/arena-human-preference-140k user prompts][Multi-turn conversations generated by GLM-5 with best-of-4 selection via Qwen3-Nemotron-235B-A22B-GenRM]Synthetic Multilingual Safety from Riva-Translate-4B-Instruct-v1.1Text132,067[Safety SFT Data: Ultra][nvidia/Riva-Translate-4B-Instruct-v1.1]Synthetic Science Reasoning Effort MediumText502,722[science-reasoning-effort-medium-v0]UndisclosedSynthetic Telecom Tool-Use Trajectories from gpt-oss-120bText12,455[Existing Tau2 telecom trajectories originally generated with DeepSeek V3.2][gpt-oss-120b]Synthetic Terminal Bench Data from OpenReasoningv2TextUndisclosed[OpenCodeReasoningv2]; [OpenMathReasoning]; [nemo-swe-bench-repos]; [SWE-Rebench]; [SWE-Fixer-110K][OpenReasoningv2]Synthetic Tulu Instruction Following from DeepSeek-R1-0528Text105,361[Nemotron-X training datasets][DeepSeek-R1-0528]Synthetic Instruction Following from gpt-oss-120bText151,988[IFEval]; [IFEvalG][gpt-oss-120b]Synthetic Instruction Following for RLTextUndisclosed[WildChat-1M]; [LMSYS-340B-Eval Dataset]; [LMSYS-Chat-1M Prompts]; [IFEval]; [IFEvalG][Qwen/Qwen3-235B-A22B-Thinking-2507]; [gpt-oss-120b]; [Qwen3-235B-A22B-Instruct-2507]Synthetic Identity Data from Qwen3-Next-80B-A3B-Instruct and Qwen3-235B-A22B-Instruct-2507Text25,992[Hand-written prompts][Qwen3-Next-80B-A3B-Instruct]; [Qwen3-235B-A22B-Instruct-2507]Synthetic Terminus Ultra Agentic Reasoning BlendText96,881[ARC-AGI-2]; [OpenCodeReasoningv2]; [OpenMathReasoning]; [SWE-Fixer-110K]; [SWE-Rebench]; [SWE-Smith][DeepSeek-V3.2]; [Qwen3-235B-A22B-Thinking-2507]; [Ring-1T]; [Kimi-K2.5]; [GLM-4.7-FP8]; [Qwen3-Next-80B-A3B-Thinking]; [gpt-oss-120b]; [Ministral-3-14B-Reasoning-2512]; [LM-4.5-Air-FP8]Synthetic STEM from Qwen3-235B-A22B-Thinking-2507Text1,174,694[IChO-IPhO-RL-v2]; [Physics-Big Dataset]; Scale HLE; [OpenMathReasoning]; [OpenCodeReasoning][Qwen3-235B-A22B-Thinking-2507]Synthetic STEM from Qwen3-235B-A22B-Instruct-2507 and gpt-oss-120bTextUndisclosed[arXiv]; [National Institutes of Health ExPorter]; [BioRxiv]; [PMC Article]; [USPTO Backgrounds]; [peS2o]; Global Regulation; [CORE]; [PG-19]; [DOAB CC BY & CC BY-SA subset]; [NDLTD][Qwen3-235B-A22B-Instruct-2507]; [gpt-oss-120b]Translation Data from TAUSText1,618,055[TAUS proprietary dataset]UndisclosedSynthetic Art of Problem Solving and Stack Exchange from gpt-oss-120b, Qwen2.5-32B-Instruct, and Goedel-Prover-V2-32BText860,469[Nemotron-Math-Proofs-v1][Goedel-Prover-V2-32B]Synthetic Art of Problem Solving and Stack Exchange from gpt-oss-120bText1,201,815[Upstream released math dataset]; [AoPS]; [StackOverflow / StackExchange][gpt-oss-120b]Synthetic Multilingual Science and Code data from DeepSeek-R1, DeepSeek-R1-0528, Qwen2.5-32B-Instruct, and Qwen3-235B-A22B, translated with Qwen2.5-32B-Instruct and Qwen2.5-14B-InstructTextUndisclosed[Nano-V3 SFT Data (without tool call)][Qwen/Qwen2.5-14B-Instruct]; [Qwen/Qwen3-4B-Thinking-2507]Synthetic Multilingual Science and Code data from DeepSeek-R1, DeepSeek-R1-0528, Qwen2.5-32B-Instruct, and Qwen3-235B-A22B, translated with Qwen2.5-32B-Instruct and Qwen2.5-14B-Instruct (Stack Exchange lineage)TextUndisclosed[Stack Exchange]; [SCP-116K]; [LIMO]; [TACO]; Code Contest; Codeforces[DeepSeek-R1]; [DeepSeek-R1-0528]; [Qwen2.5-32B-Instruct]; [Qwen3-235B-A22B]Synthetic Search Graph WalkText6,977[Wikidata / Wikipedia KnowledgeBase][MiniMaxAI/MiniMax-M2]Synthetic Agentic Diverse DomainsText281,537[Handwritten prompts (synthetic; no external seed data used)][SDG model: deepseek-ai/DeepSeek-V3.2, deepseek-ai/DeepSeek-R1-0528, Qwen/Qwen3-235B-A22B-Thinking-2507, Qwen/Qwen3-32B]; [Filtering model: openai/gpt-oss-120b, Qwen/Qwen3-32B, Qwen/Qwen3-235B-A22B-Instruct-2507]Synthetic Long Context from Qwen3-235B-A22B-Instruct-2507TextUndisclosed[Long-context SFT seed blend (pre-training blend + nano-v1 post-training data)]; [Long-context SFT data: lc_nothink 256k, MRCR 200k, RULER 256k]; [AALCR seed blend: SEC Filings, CC, Wikipedia, FinePDFs, ArXiv, Pile-NIH ExPorter, BioRxiv, PMC Article, USPTO Backgrounds, peS2o, Global Regulations, CORE, Gutenberg (PG-19), DOAB CC-BY, NDLTD, Amps, StackExchange, MathPile, Numinas][Qwen/Qwen3-235B-A22B-Thinking-2507]; [deepseek-ai/DeepSeek-R1]; [Qwen3-30B-A3B]Synthetic Nemotron Math SFT from DeepSeek-V3.2-SpecialeText1,900,553[Nemotron-Math-v2 (AOPS and StackExchange-math problems)][DeepSeek-V3.2-Speciale]Synthetic Nemotron Math TIR from DeepSeek-V3.2Text1,789,258[Nemotron-Math-v2 (AOPS and StackExchange-math problems)][DeepSeek-V3.2]Synthetic NemoCascade OCR Distillation from gpt-oss-120bText682,864[Nemotron-X training datasets][gpt-oss-120b]Synthetic CUDA 100kText93,086[KernelBook]; [HuggingFace Transformers]; [FlashInfer][gpt-oss-120b]; [DeepSeek-R1-0528]Synthetic Science MCQ and QA Diversity from GPT-OSS and Kimi-K2Text30,358[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][GPT-OSS]; [Kimi-K2]Synthetic Science HLE with Python from GPT-OSS and Kimi-K2Text85,184[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][GPT-OSS]; [Kimi-K2]Synthetic Science Search and Python from GPT-OSS and Kimi-K2Text6,179[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][GPT-OSS]; [Kimi-K2]Synthetic Science Search from GPT-OSS and Kimi-K2Text32,554[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][GPT-OSS]; [Kimi-K2]Synthetic Finance Reasoning from GPT-OSS-120B and Qwen3-235B-A22B-Instruct-2507Text326,700[SEC filings][GPT-OSS-120B, Qwen3-235B-A22B-Instruct-2507]Synthetic Science Diversity MCQ from GPT-OSS and Kimi-K2Text532,942[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][GPT-OSS]; [Kimi-K2]Synthetic Science Diversity OpenQ from GPT-OSS and Kimi-K2Text131,045[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][GPT-OSS]; [Kimi-K2]Synthetic Science Reasoning No-Tool from GPT-OSS and Kimi-K2Text2,085,600[doubtnut]; [Pile-FreeLaw]; [Llama Nemotron Dataset]; [askfilo]; [EssentialAI/essential-web-v1.0]; [Vedantu]; [auxiliary_train]; [cdquestions.com]; [AMC 8 Problems and Solutions, AMC 10 Problems and Solution, and AIME Problems and Solutions]; [AAPT]; [ICHO-IPH0 Dataset]; [LIMO dataset (Less is More for Reasoning)][GPT-OSS]; [Kimi-K2]Synthetic Text-To-SQL from gpt-oss-120bTextUndisclosed[In-house Text-to-SQL data][gpt-oss-120b]Synthetic Tool Call Schema for RL (extended)Text707,967[UltraTool]; [ToolEyes]; [AutoTools]; [API-Bank]; [Nemotron-Personas-USA]; [Salesforce xLAM function-calling]; [Glaive function-calling-v2]; [Agent-Ark/Toucan-1.5M][DeepSeek-V3.2]; [GLM-4.6]; [gpt-oss-120b]; [Kimi-K2-Instruct]Synthetic Safety from gemma-3-4b-it, Nemotron-Nano-9B-v2, and gpt-oss-120bText44,091[Safety SFT Data][google/gemma-3-4b-it]; [Nemotron-Nano-9B-v2]; [gpt-oss-120b]Synthetic Safety from DeepSeek-R1-0528, gpt-oss-120b, DeepSeek-R1-Distill-Qwen-7B, and Mixtral-8x7B-v0.1TextUndisclosed[Nemotron Content Safety Dataset V2]; [Gretel Synthetic Safety Alignment Dataset]; [RedTeam-2K]; [Malicious Tasks]; [Nemotron-Personas-USA][DeepSeek-R1-0528]; [gpt-oss-120b]; [DeepSeek-R1-Distill-Qwen-7B]; [Qwen3-30B-A3B-Thinking-2507]; [Qwen3-235B-A22B-Instruct-2507]; [Mixtral-8x7B-v0.1]Synthetic Tool Calling from Qwen3-235B-A22B-Thinking-2507 and Qwen3-Next-80B-A3B-ThinkingTextUndisclosed[ToolBench]; [glaive-function-calling-v2]; [APIGen Function-Calling]; [Nemotron-Personas-USA][Qwen3-235B-A22B-Thinking-2507]; [Qwen3-Next-80B-A3B-Thinking]Synthetic Chat from gpt-oss-120b, Mixtral-8x22B-Instruct-v0.1, Qwen3-235B-A22B-Instruct-2507, and Qwen3-235B-A22B-Thinking-2507TextUndisclosed[C4]; [LMSYS-Chat-1M]; [ShareGPT]; [GSM8K]; [PRM800K]; [FinQA]; [WikiTableQuestions]; [Riddles]; [glaive-function-calling-v2]; [SciBench]; [tigerbot-kaggle-leetcodesolutions-en-2k]; [OpenBookQA]; [Advanced Reasoning Benchmark]; Software Heritage; [Khan Academy Math Keywords]; [WildChat-1M]; [Nemotron-Personas-USA][gpt-oss-120b]; [Mixtral-8x22B-Instruct-v0.1]; [Qwen3-235B-A22B-Instruct-2507]; [Qwen3-235B-A22B-Thinking-2507]Synthetic Tool Use Interactive Agent from gpt-oss-120b, DeepSeek-R1-0528, Qwen3-32B, and Qwen3-235B-A22B-Thinking-2507TextUndisclosedNVIDIA Internal[gpt-oss-120b]; [DeepSeek-R1-0528]; [Qwen3-32B]; [Qwen3-235B-A22B-Thinking-2507]Synthetic DocFinQA and SWE-smith from Qwen3-Coder-480B-A35B-Instruct and Kimi-K2-ThinkingTextUndisclosed[DocFinQA]; [SWE-smith][Qwen3-Coder-480B-A35B-Instruct]; [Kimi-K2-Thinking]Synthetic SWE-Gym from Qwen3-Coder-480B-A35B-InstructTextUndisclosed[SWE-Gym][Qwen3-Coder-480B-A35B-Instruct]Synthetic SWE-Gym and R2E-Gym-Subset from Qwen3-Coder-480B-A35B-InstructTextUndisclosed[SWE-Gym]; [R2E-Gym-Subset][Qwen3-Coder-480B-A35B-Instruct]Synthetic SWE-Gym and R2E-Gym-Subset from DeepSeek-R1-0528TextUndisclosed[SWE-Gym]; [R2E-Gym-Subset][DeepSeek-R1-0528]Synthetic HelpSteer, LMSYS-Chat-1M, and Nemotron-Personas-USA from gpt-oss-120b, Qwen3-235B-A22B-Instruct-2507, and Qwen3-235B-A22B-Thinking-2507TextUndisclosed[HelpSteer2]; [HelpSteer3]; [LMSYS-Chat-1M]; [Nemotron-Personas-USA][gpt-oss-120b]; [Qwen3-235B-A22B-Instruct-2507]; [Qwen3-235B-A22B-Thinking-2507]Synthetic Nemotron-Personas-USA from gpt-oss-120b and Qwen3-8BTextUndisclosed[Nemotron-Personas-USA][gpt-oss-120b]; [Qwen3-8B]Vendor Terminal Bench-like Tasks (Droid)TextUndisclosed[Droid Harness Pivot vendor data]UndisclosedSynthetic BabyVisionQA Reasoning — Combined SFTText and images135,207 recordsProcedurally generated BabyVision-style visual-reasoning imagesQwen3.5-397B-A17B-FP8Synthetic BabyVision Tool-Calling SFT, Relative-Path-Corrected VariantText and images81,494 recordsNVIDIA internal BabyVision tool-calling prompts and mediaQwen3.5-397B-A17B-FP8; Kimi-K2.6Synthetic Chinese OCR v5, NMH5RText and images29,108 recordsWebster’s Unabridged Dictionary; rendered Chinese text mixed with English wordsUndisclosed; synthetic text rendering is describedSynthetic Chinese OCR v5, Augmented and CleanedText and images28,576 recordsWebster’s Unabridged Dictionary; augmented synthetic Chinese OCR imagesUndisclosed; synthetic text rendering is describedSynthetic Multimodal Safety, Energon FormatText and images22,560 recordsNVIDIA Safety-team text prompts and safe responsesEdify (image generation)Synthetic Multimodal Safety, NMH5R and Safety-FilteredText and images22,541 recordsMultimodal safety records with Edify-generated imagesEdify (image generation)Synthetic ASCII OCR v5, NMH5RText and images14,533 recordsWebster’s Unabridged Dictionary; synthetic ASCII renderingsUndisclosed; synthetic text rendering is describedSynthetic English OCR v5, NMH5RText and images14,525 recordsWebster’s Unabridged Dictionary; synthetic English renderingsUndisclosed; synthetic text rendering is described

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#language-distribution-in-post-trainingLanguage Distribution in Post-Training

For our post-training recipe, we focused on the following languages in addition to English: French, German, Italian, Japanese, Spanish, and Chinese. Those languages were represented in the form of multilingual reasoning and translation tasks.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#additional-training-dataAdditional Training Data

DataComposition and PurposeIndic continued pre-training100B tokens: 10B selected Indic tokens and 90B tokens from the preceding blend.Software-engineering mid-training67B tokens: whole repositories (20%), issue–pull-request pairs (41%), commit diffs (3%), build and CI logs (1%), agent traces (10%), pre-training replay (20%), and long-context data (5%). Sources are decontaminated against SWE-bench and Terminal-Bench repositories.Long-context pre-trainingA 35B-token extension to 1M context, followed by a 9B-token long-context recovery stage.Vision-language dataImage and video data from the Nemotron 3 Nano Omni training corpus, including visual question answering, charts, tables, OCR, multi-page documents, visual grounding, and video understanding. Post-Training DataComposition and PurposeAdditional synthetic vision dataChart understanding, grounded perception, multimodal reasoning, and visual tool-use trajectories, including data generated with Qwen3.5 and Kimi models.Reasoning trajectoriesMathematics and STEM responses generated with Kimi K3; mathematical proofs generated with DeepSeek V4 Pro; competitive-programming trajectories generated with GLM-5.2.Chat and safetyRefreshed chat responses from Qwen, Kimi, and Gemma models, with additional safety data.Long-context supervised data512K-context data, synthetic graph reasoning and transduction tasks, long-horizon search and coding trajectories, and documents containing up to hundreds of pages.Long-document translationChunked translations of PG19 and FinePDFs-Edu documents generated with DeepSeek V4 Flash.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#testing-datasetsTesting Datasets

**Data Collection Method:**Hybrid: automated, manually-collected, and synthetic **Labeling Method:**Hybrid: automated, manually-labeled, and synthetic **Properties:**Standard benchmarks and task-specific test suites covering reasoning, coding, tool use, instruction following, long context, and visual understanding.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#evaluation-datasetsEvaluation Datasets

**Data Collection Method:**Hybrid: automated, manually-collected, and synthetic **Labeling Method:**Hybrid: automated, manually-labeled, and synthetic **Properties:**Standard benchmarks and task-specific test suites covering reasoning, coding, tool use, instruction following, long context, and visual understanding.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#inferenceInference

• **Acceleration Engine:**PyTorch
• Test Hardware: - NVIDIA Hopper- H100 - H200 - NVIDIA Grace Blackwell- GB200 - GB300 - NVIDIA Blackwell- B200 - B300

The deployment examples above provide configurations for NVIDIA Hopper and Blackwell systems. Choose the precision, context length, and concurrency appropriate for the application and available memory.

https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#ethical-considerationsEthical Considerations

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.

We advise against circumvention of any provided safety guardrails contained in the Model without a substantially similar guardrail appropriate for your use case. For more details:SafetyandExplainabilitysubcards. For more detailed information on ethical considerations for this model, please see the Model Card++Bias, andPrivacysubcards.

Please report model quality, risk, security vulnerabilities or NVIDIA AI Concernshere.

Model tree fornvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16https://huggingface.co/docs/hub/model-cards#specifying-a-base-model