# Can the RTX 4060 run GroundAnything-VLM?

> Yes. The RTX 4060 runs GroundAnything-VLM fully in VRAM at Q8_0 (about 6.7 GB of 8.0 GB), at roughly 29 tokens/s.

GeForce RTX 4060: 8 GB (8 GB usable), 272 GB/s. Estimates at an 8K context.

## Every quantisation

| Quant | Quality | ≈ GB | Fit | Est. tok/s |
|---|---|---|---|---|
| Q2_K | Heavy loss | 3.18 | fits | 94 |
| Q3_K_M | Noticeable loss | 3.94 | fits | 63 |
| Q4_K_M | Small loss | 4.29 | fits | 51 |
| Q5_K_M | Near-lossless | 4.99 | fits | 43 |
| Q6_K | Near-lossless | 5.53 | fits | 37 |
| Q8_0 (pick) | Lossless | 6.66 | fits | 29 |
| F16 | Lossless | 11.1 | offload | 7.8 |

## Run it

Ask about an image (transformers):

```python
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="GroundingPI/GroundAnything-VLM", dtype="auto", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
            {"type": "text", "text": "Describe this image."},
        ],
    }
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])
```


## Vision models that fit comfortably

- [Qwen/Qwen3.5-4B](https://selfhost.directory/models/qwen/qwen3.5-4b/on/rtx-4060): Q8_0 ≈5.88 GB, est. 30 tok/s
- [google/gemma-3-4b-it](https://selfhost.directory/models/google/gemma-3-4b-it/on/rtx-4060): Q8_0 ≈5.2 GB, est. 33 tok/s
- [Qwen/Qwen3-VL-4B-Instruct](https://selfhost.directory/models/qwen/qwen3-vl-4b-instruct/on/rtx-4060): Q8_0 ≈5.82 GB, est. 32 tok/s
- [google/translategemma-4b-it](https://selfhost.directory/models/google/translategemma-4b-it/on/rtx-4060): Q8_0 ≈5.6 GB, est. 28 tok/s
- [google/medgemma-1.5-4b-it](https://selfhost.directory/models/google/medgemma-1.5-4b-it/on/rtx-4060): Q8_0 ≈5.6 GB, est. 33 tok/s
- [google/medgemma-4b-it](https://selfhost.directory/models/google/medgemma-4b-it/on/rtx-4060): Q8_0 ≈5.6 GB, est. 33 tok/s

## Other GPUs that run it

- [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
- [Arc A770 16GB](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/arc-a770): F16, est. 32 tok/s
- [Radeon RX 7800 XT](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/rx-7800-xt): F16, est. 35 tok/s
- [Tesla P40](https://selfhost.directory/models/groundingpi/groundanything-vlm/on/tesla-p40): F16, est. 20 tok/s

---
Model: https://selfhost.directory/models/groundingpi/groundanything-vlm. GPU: https://selfhost.directory/models/gpus/rtx-4060. Source: https://selfhost.directory/models/groundingpi/groundanything-vlm/on/rtx-4060