LLM EXPLORER 61,836 MODELS INDEXED

LocateAnything 3B by nvidia

By nvidia · 1136916 downloads
HF Repo ↗ Copy URL

LocateAnything 3B is an open-source language model by nvidia. Features: 3b LLM, VRAM: 7.7GB, License: other, Instruction-Based, LLM Explorer Score: 0.46.

  Arxiv:1405.0312   Arxiv:1908.03195   Arxiv:2109.10852   Arxiv:2303.05499   Arxiv:2504.07491   Arxiv:2504.07981   Arxiv:2510.12798   Arxiv:2605.27365 Base model:finetune:qwen/qwen2... Base model:qwen/qwen2.5-3b-ins...   Conversational   Custom code   Eagle   En   Feature-extraction   Grounding   Image-text-to-text   Instruct   Locateanything   Nvidia   Object-detection   Region:us   Safetensors   Sharded   Tensorflow   Vision
Model Card on HF 🤗: https://huggingface.co/nvidia/LocateAnything-3B 

LocateAnything 3B Parameters and Internals

LLM NameLocateAnything 3B
Repository 🤗https://huggingface.co/nvidia/LocateAnything-3B 
Base Model(s)  Qwen/Qwen2.5-3B-Instruct   Qwen/Qwen2.5-3B-Instruct
Model Size3b
Required VRAM7.7 GB
Updated2026-07-30
Maintainernvidia
Model Typelocateanything
Instruction-BasedYes
Model Files  5.0 GB: 1-of-2   2.7 GB: 2-of-2   0.0 GB
Supported Languagesen
Model ArchitectureLocateAnythingForConditionalGeneration
Licenseother
Model Max Length16384
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Torch Data Typebfloat16
Errorsreplace

Quantized Models of the LocateAnything 3B

Model
Likes
Downloads
VRAM
LocateAnything 3B 4bit204353 GB
LocateAnything 3B 8bit101934 GB