NVIDIA
Explore
Models
Blueprints
GPUs
Docs
⌘KCtrl+K
Terms of Use
Privacy Policy
Your Privacy Choices
Contact

Copyright © 2026 NVIDIA Corporation

Models

Deploy and scale models on your GPU infrastructure of choice with NVIDIA NIM inference microservices

Optimized by NVIDIALaunch from Hugging FaceBeta

Filters (1)

  • Download Available
    13
  • API Endpoint
    5
  • Code Generation
    2
  • Retrieval Augmented Generation
    0
  • Drug Discovery
    0
  • Image-to-Text
    0
  • Object Detection
    0
  • NVIDIA
    5
  • DeepSeek AI
    3
  • Mistral AI
    2
  • OpenAI
    2
  • Meta
    1
  • Reasoning
    • 18 models
    OpenAI

    gpt-oss-20b

    Smaller Mixture of Experts (MoE) text-only LLM for efficient AI reasoning and math
    text-to-text
    7.06M
    7mo
    OpenAI

    gpt-oss-120b

    Mixture of Experts (MoE) reasoning LLM (text-only) designed to fit within 80GB GPU.
    text-to-text
    34.11M
    7mo
    NVIDIA

    llama-3.3-nemotron-super-49b-v1.5

    High efficiency model with leading accuracy for reasoning, tool calling, chat, and instruction following.
    chat
    4.24M
    7mo
    Sarvamai

    sarvam-m

    Multilingual, hybrid-reasoning model optimized for Indian language tasks, programming, mathematical reasoning capabilities.
    coding
    421K
    7mo
    Microsoft

    phi-4-mini-flash-reasoning

    Lightweight reasoning model for applications in latency bound, memory/compute constrained environments
    edge
    413K
    7mo
    Mistral AI

    magistral-small-2506

    High performance reasoning model optimized for efficiency and edge deployment
    coding
    3.41M
    7mo
    NVIDIA

    llama-3.1-nemotron-nano-4b-v1.1

    State-of-the-art open model for reasoning, code, math, and tool calling - suitable for edge agents
    edge
    98.6K
    8mo
    Marin

    marin-8b-instruct

    State-of-the-art open model trained on open datasets, excelling in reasoning, math, and science.
    Reasoning
    435K
    9mo
    NVIDIA

    llama-3.1-nemotron-ultra-253b-v1

    Superior inference efficiency with highest accuracy for scientific and complex math reasoning, coding, tool calling, and instruction following.
    chat
    6.62M
    7mo
    Qwen

    qwq-32b

    Powerful reasoning model capable of thinking and reasoning, can achieve significantly enhanced performance in downstream tasks, especially hard problems.
    coding
    3.32M
    8mo
    NVIDIA

    llama-3.3-nemotron-super-49b-v1

    High efficiency model with leading accuracy for reasoning, tool calling, chat, and instruction following.
    chat
    1.07M
    7mo
    NVIDIA

    llama-3.1-nemotron-nano-8b-v1

    Leading reasoning and agentic AI accuracy model for PC and edge.
    chat
    552K
    8mo
    DeepSeek AI

    deepseek-r1-distill-llama-8b

    Distilled version of Llama 3.1 8B using reasoning data generated by DeepSeek R1 for enhanced performance.
    Distillation
    4.14M
    8mo
    DeepSeek AI

    deepseek-r1-distill-qwen-32b

    Distilled version of Qwen 2.5 32B using reasoning data generated by DeepSeek R1 for enhanced performance.
    coding
    2.44K4.17M
    9mo
    DeepSeek AI

    deepseek-r1-distill-qwen-14b

    Distilled version of Qwen 2.5 14B using reasoning data generated by DeepSeek R1 for enhanced performance.
    coding
    2.07K3.78M
    9mo
    Mistral AI

    mistral-small-24b-instruct

    Latency-optimized language model excelling in code, math, general knowledge, and instruction-following.
    code
    531K
    8mo
    Tiiuae

    falcon3-7b-instruct

    Instruction tuned LLM achieving SoTA performance on reasoning, math and general knowledge capabilities
    Coding
    489K
    9mo
    Meta

    llama-3.3-70b-instruct

    Advanced LLM for reasoning, math, general knowledge, and function calling
    Reasoning
    23.42M
    8mo
    Items per page
    of 1 pages