Nemotron-3.5-Rerank-8B-BF16
Nemotron-3.5-Rerank-8B-BF16 is an 8.38B-parameter reranker that scores the relevance of a document to a text query. It accepts text, images, or combined text and images as documents, and is described as supporting multilingual and multimodal retrieval.
What it does
Use it to rerank candidate text or image-containing documents in a retrieval pipeline, including vision RAG systems.
- Availability
- Released on Hugging Face 10/20/2026
- Pricing model
- Not established
- Deployment
- Global; designed and/or optimized for NVIDIA GPU-accelerated systems
- Privacy
- Not established
- Licence
- OpenMDW License Agreement, version 1.1 (OpenMDW-1.1); base model governed by Apache License 2.0
Capability
Modality
Access
Deployment
Pricing
Target user
Evidence
6 claims- The model produces a logit representing the relevance of a document to a text query. [1]
It is a world-class reranker optimized to produce a logit representing the relevance of a document to a text query.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - Documents may contain text, page images, or both. [1]
Documents may contain text, page images, or both, including charts, tables, and infographics.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model was evaluated for text retrieval across 34 languages. [1]
For text retrieval, the model was evaluated across 34 languages: English, Arabic, Assamese, Bengali, Bulgarian, Chinese, Danish, Dutch, Finnish, French, German, Hindi, Hinglish, Indonesian, Italian, Japanese, Korean, Malay, Marathi, Nepali, Norwegian, Persian, Portuguese, Romanian, Russian, Spanish, Swahili, Swedish, Tamil, Telugu, Thai, Ukrainian, Urdu, and Vietnamese.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model was evaluated for image retrieval across 18 languages. [1]
For text retrieval, the model was evaluated across 34 languages: English, Arabic, Assamese, Bengali, Bulgarian, Chinese, Danish, Dutch, Finnish, French, German, Hindi, Hinglish, Indonesian, Italian, Japanese, Korean, Malay, Marathi, Nepali, Norwegian, Persian, Portuguese, Romanian, Russian, Spanish, Swahili, Swedish, Tamil, Telugu, Thai, Ukrainian, Urdu, and Vietnamese.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model was released on Hugging Face on 10/20/2026. [1]
Set `MODEL_PATH` to the Hugging Face model ID or a local checkpoint directory:
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model is ready for commercial or non-commercial use. [1]
This model is ready for commercial or non-commercial use.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights.
Sources
1
1
Hugging FaceOwner-controlled
https://huggingface.co/nvidia/Nemotron-3.5-Rerank-8B-BF16
Retrieved 2026-10-29 00:46 UTC
Verified
Versions
4| Version | Effective | Change | Changed fields | State |
|---|---|---|---|---|
| 4 | 2026-10-21 12:15 UTC | material change | access, availability, deployment | Current |
| 3 | 2026-10-19 14:01 UTC | material change | access, availability, deployment | Current |
| 2 | 2026-10-17 15:52 UTC | material change | availability, deployment | Current |
| 1 | 2026-10-16 16:41 UTC | publication | None | Current |
In briefs
1- One model endpoint for chat, speech, decisions, and safety checks Daily brief 2026-10-16
Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.