LocateAnything-3B
Support summary
LocateAnything-3B is a checkpoint published by NVIDIA, built on the LocateAnythingForConditionalGeneration architecture. SGLang 0.5.19, the newest indexed SGLang release, documents support.
SGLang documents support in 6 of 13 indexed releases, from 0.5.15 to 0.5.19.
- Configured context
- 32,768 tokens
- License
- nvidia-license
Hugging Face model card
- Publisher
- NVIDIA
- Task
- image text to text
- License
- nvidia-license
- Library
- transformers
- Architecture
- LocateAnythingForConditionalGeneration
- Weight tensor elements
- 3.8B
- Tensor types
- BF16
- Configured context
- 32,768 tokens
- Languages
- en
- Base models
- Qwen/Qwen2.5-3B-Instruct
- Weight access
- Public
- Repository created
- 2026-03-02
Model notes and sources
- SGLang v0.5.19 · docs/docs/supported-models/multimodal_language_models.mdx · 3 indexed revisions, v0.5.17 to v0.5.19
- SGLang v0.5.16 · docs_new/docs/supported-models/multimodal_language_models.mdx · 3 indexed revisions, v0.5.15 to v0.5.16
- nvidia/LocateAnything-3B · publisher model card
- nvidia/LocateAnything-3B · Hugging Face metadata
- nvidia/LocateAnything-3B · publisher configuration
- Reviewed
Engine support
3 engines
| Engine | Release | Checkpoint support | Evidence scope | Last known supported |
|---|---|---|---|---|
| vLLM | 0.28.0Newest indexed | No matching evidence curated | No supported release indexed | |
Features, dependencies and evidence for vLLM 0.28.0 | ||||
| SGLang | 0.5.19Newest indexed | Checkpoint evidenceSource: SGLang v0.5.19 · docs/docs/supported-models/multimodal_language_models.mdxReviewed | 0.5.19This release | |
Features, dependencies and evidence for SGLang 0.5.19 | ||||
| TensorRT-LLM | 1.2.1Newest indexed | No matching evidence curated | No supported release indexed | |
Features, dependencies and evidence for TensorRT-LLM 1.2.1 | ||||