selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

GroundingPI logo

Can the RTX 3060 12GB run GroundAnything-VLM?

GroundAnything-VLM·4.8B·GeForce RTX 3060 12GB·12 GB VRAM

Saved in this browser, used on every model page.
Yes

Yes. The RTX 3060 12GB runs GroundAnything-VLM fully in VRAM at FP16 / BF16 (about 11 GB of 12 GB), at roughly 20 tokens/s.

best quant
FP16 / BF16
needs
11 GB
usable
12 GB
est. speed
~20 tok/s

Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the GroundAnything-VLM page →

Every quantisation of GroundAnything-VLM on the RTX 3060 12GB

QuantQualityMemoryFitSpeed
Q2_K Heavy loss3.2 GBfits~124 tok/s
Q3_K_M Noticeable loss3.9 GBfits~84 tok/s
Q4_K_M Small loss4.3 GBfits~67 tok/s
Q5_K_M Near-lossless5.0 GBfits~57 tok/s
Q6_K Near-lossless5.5 GBfits~50 tok/s
Q8_0 Lossless6.7 GBfits~38 tok/s
FP16 / BF16 pickLossless11 GBfits~20 tok/s

Speed is a bandwidth-based estimate for generating text (360 GB/s); prompt processing and long contexts are slower.

Measured speed on the RTX 3060 12GB

No community reports for this pair yet.

Report your own speed →

Run GroundAnything-VLM on the RTX 3060 12GB

Transformers · Ask about an imagedocs
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 the RTX 3060 12GB comfortably

All that fit →
ModelParamsBest quantMemorySpeedLicence
4.7BQ8_0≈5.9 GBFits in memory~40 tok/sCommercial OK
4.3BQ8_0≈5.2 GBFits in memory~43 tok/sConditions
4.4BQ8_0≈5.8 GBFits in memory~42 tok/sCommercial OK
5BQ8_0≈5.6 GBFits in memory~37 tok/sConditions
4.3BQ8_0≈5.6 GBFits in memory~43 tok/sConditions
3.8BQ8_0≈4.0 GBFits in memory~50 tok/sNon-commercial

Other GPUs that run GroundAnything-VLM

Frequently asked questions

Can the RTX 3060 12GB run GroundAnything-VLM?
Yes. The RTX 3060 12GB runs GroundAnything-VLM fully in VRAM at FP16 / BF16 (about 11 GB of 12 GB), at roughly 20 tokens/s.
How fast does GroundAnything-VLM run on the RTX 3060 12GB?
An estimate from memory bandwidth gives about 20 tokens/s at FP16 / BF16.
Which quantisation of GroundAnything-VLM should I use on the RTX 3060 12GB?
FP16 / BF16: the largest one that fits. Expected quality: lossless compared with the full-precision weights.

Memory and speed are estimates from the model's size and the card's memory bandwidth; real numbers depend on the runtime, drivers and context length. Community figures are medians of user reports.