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.

GrootTeam logo

Can the Apple M4 (16 GB) run Groot-19B-A3B-261023?

Groot-19B-A3B-261023·19.4B (3B active)·Apple M4 (16 GB)·16 GB unified memory

Saved in this browser, used on every model page.
Just about

Just about. At Q3_K_M it needs about 11 GB of the 11 GB available, so keep the context short, at roughly 45 tokens/s.

best quant
Q3_K_M
needs
11 GB
usable
11 GB
est. speed
~45 tok/s

Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the Groot-19B-A3B-261023 page →

Every quantisation of Groot-19B-A3B-261023 on the Apple M4 (16 GB)

QuantQualityMemoryFitSpeed
Q2_K Heavy loss8.8 GBfits~67 tok/s
Q3_K_M pickNoticeable loss11 GBtight~45 tok/s
Q4_K_M Small loss13 GBtoo big—
Q5_K_M Near-lossless16 GBtoo big—
Q6_K Near-lossless18 GBtoo big—
Q8_0 Lossless21 GBtoo big—
FP16 / BF16 Lossless38 GBtoo big—

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 Groot-19B-A3B-261023 on the Apple M4 (16 GB)

Using Q3_K_M, the best fit for Apple M4 (16 GB).

Ollama · Q3_K_M straight from Hugging Facedocs
ollama run hf.co/bartowski/GrootTeam_Groot-19B-A3B-261023-GGUF:Q3_K_M
llama.cpp · OpenAI-compatible server with a web UIdocs
llama-server -hf bartowski/GrootTeam_Groot-19B-A3B-261023-GGUF:Q3_K_M

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

LM Studio · Download with the lms CLIdocs
lms get bartowski/GrootTeam_Groot-19B-A3B-261023-GGUF@Q3_K_M

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="GrootTeam/Groot-19B-A3B-261023", 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
9.7BQ6_K≈8.8 GBFits in memory~8.3 tok/sCommercial OK
8.8BQ8_0≈10 GBFits in memory~7.1 tok/sCommercial OK
10.7BQ5_K_M≈9.2 GBFits in memory~8.7 tok/sConditions
9.4BQ6_K≈9.1 GBFits in memory~8.5 tok/sCommercial OK
8.3BQ8_0≈8.9 GBFits in memory~7.5 tok/sCommercial OK
10.3BQ6_K≈8.9 GBFits in memory~7.8 tok/sCommercial OK

Other GPUs that run Groot-19B-A3B-261023

Frequently asked questions

Can the Apple M4 (16 GB) run Groot-19B-A3B-261023?
Just about. At Q3_K_M it needs about 11 GB of the 11 GB available, so keep the context short, at roughly 45 tokens/s.
How fast does Groot-19B-A3B-261023 run on the Apple M4 (16 GB)?
An estimate from memory bandwidth gives about 45 tokens/s at Q3_K_M.
Which quantisation of Groot-19B-A3B-261023 should I use on the Apple M4 (16 GB)?
Q3_K_M: the largest one that fits. Expected quality: noticeable loss 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.