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.

LiquidAI logo

Can the Apple M4 (16 GB) run LFM3-VL-2B?

LFM3-VL-2B·1.9B·Apple M4 (16 GB)·16 GB unified memory

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

Yes. The Apple M4 (16 GB) runs LFM3-VL-2B fully in unified memory at FP16 / BF16 (about 4.3 GB of 11 GB), at roughly 17 tokens/s.

best quant
FP16 / BF16
needs
4.3 GB
usable
11 GB
est. speed
~17 tok/s

Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the LFM3-VL-2B page →

Every quantisation of LFM3-VL-2B on the Apple M4 (16 GB)

QuantQualityMemoryFitSpeed
Q2_K Heavy loss1.6 GBfits~105 tok/s
Q3_K_M Noticeable loss1.7 GBfits~70 tok/s
Q4_K_M Small loss1.9 GBfits~57 tok/s
Q5_K_M Near-lossless2.1 GBfits~48 tok/s
Q6_K Near-lossless2.3 GBfits~42 tok/s
Q8_0 Lossless2.7 GBfits~32 tok/s
FP16 / BF16 pickLossless4.3 GBfits~17 tok/s

Speed is a bandwidth-based estimate for generating text (120 GB/s); prompt processing and long contexts are slower.

Measured speed on the Apple M4 (16 GB)

No community reports for this pair yet.

Report your own speed →

Run LFM3-VL-2B on the Apple M4 (16 GB)

Using FP16 / BF16, the best fit for Apple M4 (16 GB).

Ollama · From the Ollama librarydocs
ollama run lfm3-vl:2b

The library build picks its own default quantisation.

Ollama · F16 straight from Hugging Facedocs
ollama run hf.co/LiquidAI/LFM3-VL-2B-GGUF:F16
llama.cpp · OpenAI-compatible server with a web UIdocs
llama-server -hf LiquidAI/LFM3-VL-2B-GGUF:F16

brew install llama.cpp ships with Metal. Downloads the GGUF on first run.

LM Studio · Download with the lms CLIdocs
lms get LiquidAI/LFM3-VL-2B-GGUF@F16

Then load it from the LM Studio app or with `lms load`.

Transformers · Ask about an imagedocs
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="LiquidAI/LFM3-VL-2B", 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 Apple M4 (16 GB) comfortably

All that fit →
ModelParamsBest quantMemorySpeedLicence
LiquidAI logod1-2BLiquidAI
1.9BQ8_0≈2.7 GBFits in memory~32 tok/sConditions
2.3BQ8_0≈2.8 GBFits in memory~27 tok/sCommercial OK
2.1BQ8_0≈3.2 GBFits in memory~29 tok/sCommercial OK
1.6BQ8_0≈1.9 GBFits in memory~40 tok/sCommercial OK
1.3BQ8_0≈1.7 GBFits in memory~48 tok/sCommercial OK
2.2BQ8_0≈2.6 GBFits in memory~28 tok/sCommercial OK

Other GPUs that run LFM3-VL-2B

Frequently asked questions

Can the Apple M4 (16 GB) run LFM3-VL-2B?
Yes. The Apple M4 (16 GB) runs LFM3-VL-2B fully in unified memory at FP16 / BF16 (about 4.3 GB of 11 GB), at roughly 17 tokens/s.
How fast does LFM3-VL-2B run on the Apple M4 (16 GB)?
An estimate from memory bandwidth gives about 17 tokens/s at FP16 / BF16.
Which quantisation of LFM3-VL-2B should I use on the Apple M4 (16 GB)?
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.