# GroundAnything-VLM

> Model family: GroundAnything — DLM / parallel decoding · GroundAnything-VLM — autoregressive.

- **Organisation:** GroundingPI
- **Category:** Vision-Language Models, OCR & Document AI
- **Parameters:** 4.8B
- **Context length:** 262,144 tokens
- **Licence:** apache-2.0-with-upstream-kimi-k3 (Commercial use OK)
- **Released:** 2026-09-30
- **Hugging Face:** 1.2K downloads (30 days), 69 likes

## Estimated memory

| Quant | ≈ GB |
|---|---|
| Q2_K | 3.18 |
| Q3_K_M | 3.94 |
| Q4_K_M | 4.5 |
| Q5_K_M | 4.99 |
| Q6_K | 5.53 |
| Q8_0 | 6.66 |
| F16 | 11.1 |

## Runs with

- [Transformers](https://github.com/huggingface/transformers)

## About

Model family: GroundAnything — DLM / parallel decoding · GroundAnything-VLM — autoregressive.

This repository contains the autoregressive GroundAnything-VLM checkpoint. The family overview and figures below are shared with the DLM page; use the VLM serving recipe for these weights.

• ModelScope: GroundAnything (DLM) · GroundAnything-VLM (VLM).
• 🚀 Online Demo: Hugging Face Spaces.
• 🌐 Project Page: GroundAnything.
• 💻 GitHub Code: groundingpi/GroundAnything.
• 📄 Paper: arXiv:2609.39600.
• 🤗 HF paper:

Autoregressive (AR) grounding models serialize spatial predictions, introducing sequential latency and imposing a causal order on output tokens. We view grounding as visual evidence extraction: objects, locations, and spatial relations are jointly constrained by the image and query, yet their dependencies do not imply an intrinsic left-to-right generation order. This distinction makes bidirectional diffusion a natural fit, allowing spatial hypotheses to emerge in parallel and be jointly refined through iterative denoising.

## Alternatives

- [Qwen/Qwen3.5-4B](https://selfhost.directory/models/qwen/qwen3.5-4b) (4.7B, Apache 2.0, ≈4.26 GB at Q4): Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
- [Qwen/Qwen3-VL-4B-Instruct](https://selfhost.directory/models/qwen/qwen3-vl-4b-instruct) (4.4B, Apache 2.0, ≈4.26 GB at Q4): Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.
- [Qwen/Qwen3.5-4B-Base](https://selfhost.directory/models/qwen/qwen3.5-4b-base) (4.7B, Apache 2.0, ≈4.26 GB at Q4): Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
- [tencent/WeVisOne-VL-4B](https://selfhost.directory/models/tencent/wevisone-vl-4b) (4.8B, Apache 2.0, ≈4.49 GB at Q4): WeVisOne-VL (WeChat Vision All-in-One) is a unified vision-language model developed by WeChat Vision, Tencent Inc. Built upon Qwen3-VL, it unifies object-level perception (open-vocabulary object detection, referring localization, object counting, object pointing, and OCR) and pixel-level perception…
- [h2oai/h2o-lightning-4b](https://selfhost.directory/models/h2oai/h2o-lightning-4b) (4.5B, Apache 2.0, ≈4.19 GB at Q4): H2O-Lightning-4B is a 4B-parameter decision model from H2O.ai, built on Qwen/Qwen3.5-4B. It answers typed decision questions about a record (a document, a ticket, a policy, a conversation) and about images that come with the record (photos, screenshots, scanned documents, charts).
- [google/medgemma-1.5-4b-it](https://selfhost.directory/models/google/medgemma-1.5-4b-it) (4.3B, health-ai-developer-foundations, ≈4.11 GB at Q4): * License: The use of MedGemma is governed by the Health AI Developer Foundations terms of use. MedGemma has not been evaluated or optimized for multi-turn applications.

More: https://selfhost.directory/models/groundingpi/groundanything-vlm/alternatives

## Runs well on

- [GeForce RTX 4060](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/rtx-4060): Q8_0, est. 29 tok/s
- [Apple M4 (16 GB)](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/apple-m4-16gb): Q8_0, est. 13 tok/s
- [Radeon RX 6700 XT](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/rx-6700-xt): F16, est. 22 tok/s
- [GeForce RTX 3060 12GB](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/rtx-3060): F16, est. 20 tok/s
- [Arc B580](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/arc-b580): F16, est. 26 tok/s
- [Tesla T4](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/tesla-t4): F16, est. 18 tok/s

## Similar size

- [Qwen/Qwen3.5-4B](https://selfhost.directory/models/qwen/qwen3.5-4b) (4.7B)
- [google/gemma-3-4b-it](https://selfhost.directory/models/google/gemma-3-4b-it) (4.3B)
- [Qwen/Qwen3-VL-4B-Instruct](https://selfhost.directory/models/qwen/qwen3-vl-4b-instruct) (4.4B)
- [Qwen/Qwen2.5-VL-3B-Instruct](https://selfhost.directory/models/qwen/qwen2.5-vl-3b-instruct) (3.8B)
- [google/translategemma-4b-it](https://selfhost.directory/models/google/translategemma-4b-it) (5B)
- [google/medgemma-1.5-4b-it](https://selfhost.directory/models/google/medgemma-1.5-4b-it) (4.3B)

## FAQ

**How much VRAM does GroundAnything-VLM need?**
About 4.5 GB at Q4_K_M with an 8K context. Q8_0 needs about 6.7 GB and full precision about 11 GB. Longer contexts need more memory for the KV cache.

**Can GroundAnything-VLM run on the RTX 4060?**
Yes. It fits at Q8_0 (about 6.7 GB), at an estimated 29 tokens/s. It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.

**Can I use GroundAnything-VLM commercially?**
Yes. It is released under apache-2.0-with-upstream-kimi-k3. Custom licence: read the model card before using it commercially.

**What is the context length of GroundAnything-VLM?**
262,144 tokens (256K). The full window needs far more memory than an 8K context.

**How do I run GroundAnything-VLM locally?**
It runs with Transformers. Each has copy-paste examples in its docs.

## Links

- huggingface: https://huggingface.co/GroundingPI/GroundAnything-VLM
- homepage: https://groundingpi.github.io/groundanything/
- github: https://github.com/groundingpi/GroundAnything
- paper: https://arxiv.org/abs/2609.39600
- demo: https://huggingface.co/spaces/GroundingPI/GroundAnything-VLM

---
Source: https://selfhost.directory/models/groundingpi/groundanything-vlm