NVIDIA
Llama 3.2 NeMo Retriever Multimodal Reranking 1B
Container
NVIDIA
Llama 3.2 NeMo Retriever Multimodal Reranking 1B

World-class multimodal question-answering retrieval reranking NIM.

  • LayerLabelCreated
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4CMD
    /opt/nim/start_server.sh
    07/24/2026 10:07 PM UTC
    sha256:1594490fe61a23957a9dbfd6c63faa7955aad4af0fbf83156cb9e5d57983b35dCOPY
    NOTICE /opt/nim/NOTICE
    07/24/2026 10:07 PM UTC
    sha256:10bf369d70e1d988994c464ffeb1b9056aa9bac33d75a402bc1eb730f530e395COPY
    VERSION /opt/nim/VERSION
    07/24/2026 10:07 PM UTC
    sha256:8267dee995e3c5e0d29f98bc5de031ec84369ccbb67aa6ea7211a2185f8d3e6fCOPY
    LICENSE /opt/nim/LICENSE
    07/24/2026 10:07 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NIM_ENGINE_MODEL_DOWNLOAD_PROVIDER=ngc
    07/24/2026 10:07 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NIM_PIPELINE_MAX_SEQ_LEN=8192
    07/24/2026 10:07 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NIM_ENGINE_MODEL_PATH=/model/rerank
    07/24/2026 10:07 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NIM_ENGINE_MODEL_NAME=nvidia/llama-nemotron-rerank-vl-1b-v2
    07/24/2026 10:07 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    RUST_LOG=nemotron_serving_telemetry::startup_vram=info
    07/24/2026 10:07 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NIM_HTTP_API_PORT=8000
    07/24/2026 10:07 PM UTC