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.

thinkingmachines logo

Can the Apple M3 Ultra (512 GB) run Inkling-2-Small?

Inkling-2-Small·383B·Apple M3 Ultra (512 GB)·512 GB unified memory

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

Yes. The Apple M3 Ultra (512 GB) runs Inkling-2-Small fully in unified memory at Q6_K (about 311 GB of 384 GB), at roughly 1.4 tokens/s.

best quant
Q6_K
needs
311 GB
usable
384 GB
est. speed
~1.4 tok/s

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

Every quantisation of Inkling-2-Small on the Apple M3 Ultra (512 GB)

QuantQualityMemoryFitSpeed
Q2_K Heavy loss125 GBfits~3.6 tok/s
Q3_K_M Noticeable loss185 GBfits~2.4 tok/s
Q4_K_M Small loss229 GBfits~1.9 tok/s
Q5_K_M Near-lossless268 GBfits~1.7 tok/s
Q6_K pickNear-lossless311 GBfits~1.4 tok/s
Q8_0 Lossless400 GBtoo big—
FP16 / BF16 Lossless476 GBtoo big—

Speed is a bandwidth-based estimate for generating text (819 GB/s); prompt processing and long contexts are slower.

Measured speed on the Apple M3 Ultra (512 GB)

No community reports for this pair yet.

Report your own speed →

Run Inkling-2-Small on the Apple M3 Ultra (512 GB)

Using Q4_K_M, the best fit for Apple M3 Ultra (512 GB).

Ollama · From the Ollama librarydocs
ollama run inkling-2-small

The library build picks its own default quantisation.

Ollama · UD-Q4_K_XL straight from Hugging Facedocs
ollama run hf.co/unsloth/Inkling-2-Small-GGUF:UD-Q4_K_XL
Transformers · Ask about an imagedocs
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="thinkingmachines/Inkling-2-Small", 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 M3 Ultra (512 GB) comfortably

All that fit →
ModelParamsBest quantMemorySpeedLicence
321BA18.3BQ8_0≈356 GBFits in memory~23 tok/sCommercial OK
MiniMaxAI logoMiniMax-M3MiniMaxAI
427BQ6_K≈345 GBFits in memory~1.3 tok/sConditions
403BA17BQ6_K≈321 GBFits in memory~32 tok/sCommercial OK
180BA61.6BQ8_0≈185 GBFits in memory~6.9 tok/sConditions
220BQ8_0≈3.5 GBFits in memory~1.9 tok/sConditions
305BA15.2BQ8_0≈318 GBFits in memory~28 tok/sCommercial OK

Frequently asked questions

Can the Apple M3 Ultra (512 GB) run Inkling-2-Small?
Yes. The Apple M3 Ultra (512 GB) runs Inkling-2-Small fully in unified memory at Q6_K (about 311 GB of 384 GB), at roughly 1.4 tokens/s.
How fast does Inkling-2-Small run on the Apple M3 Ultra (512 GB)?
An estimate from memory bandwidth gives about 1.4 tokens/s at Q6_K.
Which quantisation of Inkling-2-Small should I use on the Apple M3 Ultra (512 GB)?
Q6_K: the largest one that fits. Expected quality: near-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.