Llama Nemotron Rerank
NVIDIANVIDIA's NeMo Retriever reranking model, a 1B cross-encoder fine-tuned from Llama-3.2-1B with bidirectional attention and a binary relevance head. It supports 8,192-token documents and was evaluated across 26 languages on MIRACL, MLQA and MLDR, and it is the reranking stage NVIDIA ships as a NIM microservice alongside its llama-nemotron embedding model. A vision-language sibling, llama-nemotron-rerank-vl-1b-v2, extends it to visual documents.
Openness
3 medium confidence- weights
- open(ungated safetensors for the text reranker
- data
- closed(800k samples from an unnamed blend of commercially licensed public QA datasets
- code
- partial(card usage and NIM deployment only
- license
- OpenMDW-1.1(the governing licence named in the card)+Llama-3.2-Community(named as additional information
The Hub's `other` license tag resolves, in the card's own text, to two licenses at once: the permissive OpenMDW License Agreement v1.1 and the Llama 3.2 Community License, which carries a 700 million monthly-active-user cap. The more restrictive of the two governs, and confidence stays medium because the terms are split between a bare tag and a prose section rather than declared in one place.
- https://huggingface.co/nvidia/llama-nemotron-rerank-1b-v2/raw/main/README.md recorded 2026-09-11
Front matter reads license: other, license_name: openmdw-1.1, license_link https://openmdw.ai/license/1-1/. The License/Terms of use section says use is governed by the OpenMDW License Agreement version 1.1, with Additional Information: Llama 3.2 Community Model License Agreement; Built with Llama. The card states the model is ready for commercial use, describes a fine-tune of meta-llama/Llama-3.2-1B, and records training on 800k samples from an unnamed blend of public QA datasets chosen for commercial licensing.
Adoption
3 high confidenceAdoption is measured as Hugging Face downloads for llama-nemotron-rerank-1b-v2, the text reranker. Its vision-language sibling is scored as a separate product under the category's rule that a VL line is distinct from the text line, and including it would not change this reading.
- https://huggingface.co/api/models/nvidia/llama-nemotron-rerank-1b-v2 recorded 2026-10-07
724,549 downloads in the trailing 30 days for nvidia/llama-nemotron-rerank-1b-v2
Capability
3 medium confidenceLlama Nemotron Rerank has real multilingual and long-document coverage and a production NIM deployment path, but it is measured as Recall@5 inside NVIDIA's own pipeline, where an older sibling model scores higher, rather than on any public leaderboard. It sits a step below Qwen3-Reranker, which publishes BEIR and MMTEB numbers on the benchmarks this category treats as primary.
- https://huggingface.co/nvidia/llama-nemotron-rerank-1b-v2/raw/main/README.md recorded 2026-09-11
Evaluation tables report llama-nemotron-embed-1b-v2 + llama-nemotron-rerank-1b-v2 at 73.64% average Recall@5 on NQ, HotpotQA, FiQA and TechQA against 75.45% for nv-embedqa-e5-v5 + nv-rerankQA-mistral-4b-v3, and 65.80% on MIRACL multilingual; the overview states support for 8192-token documents and evaluation on 26 languages.
Verified 2026-09-11, except some scores, still awaiting re-confirmation