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.

nvidia logo

Can the Apple M4 Max 40-core (128 GB) run NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball?

NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball·33B·Apple M4 Max 40-core (128 GB)·128 GB unified memory

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

Yes. The Apple M4 Max 40-core (128 GB) runs NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball fully in unified memory at FP16 / BF16 (about 65 GB of 96 GB), at roughly 4.5 tokens/s.

best quant
FP16 / BF16
needs
65 GB
usable
96 GB
est. speed
~4.5 tok/s

Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball page →

Every quantisation of NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball on the Apple M4 Max 40-core (128 GB)

QuantQualityMemoryFitSpeed
Q2_K Heavy loss12 GBfits~28 tok/s
Q3_K_M Noticeable loss17 GBfits~19 tok/s
Q4_K_M Small loss20 GBfits~15 tok/s
Q5_K_M Near-lossless24 GBfits~13 tok/s
Q6_K Near-lossless28 GBfits~11 tok/s
Q8_0 Lossless35 GBfits~8.6 tok/s
FP16 / BF16 pickLossless65 GBfits~4.5 tok/s

Speed is a bandwidth-based estimate for generating text (546 GB/s); prompt processing and long contexts are slower.

Measured speed on the Apple M4 Max 40-core (128 GB)

No community reports for this pair yet.

Report your own speed →

Omni models that fit the Apple M4 Max 40-core (128 GB) comfortably

All that fit →
ModelParamsBest quantMemorySpeedLicence
35.3BA3BQ8_0≈33 GBFits in memory~94 tok/sCommercial OK
33BA3BQ8_0≈34 GBFits in memory~94 tok/sCommercial OK
31.7BA3BQ8_0≈33 GBFits in memory~94 tok/sCommercial OK
31.7BA3BQ8_0≈34 GBFits in memory~94 tok/sCommercial OK
24.3BQ8_0≈26 GBFits in memory~12 tok/sCommercial OK
stepfun-ai logoStep-Audio-R1stepfun-ai
33.5BQ8_0≈37 GBFits in memory~8.4 tok/sCommercial OK

Other GPUs that run NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball

Frequently asked questions

Can the Apple M4 Max 40-core (128 GB) run NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball?
Yes. The Apple M4 Max 40-core (128 GB) runs NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball fully in unified memory at FP16 / BF16 (about 65 GB of 96 GB), at roughly 4.5 tokens/s.
How fast does NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball run on the Apple M4 Max 40-core (128 GB)?
An estimate from memory bandwidth gives about 4.5 tokens/s at FP16 / BF16.
Which quantisation of NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball should I use on the Apple M4 Max 40-core (128 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.