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.

mehdi-hf logo

Can the Tesla P40 run cilla-mini?

cilla-mini·28M·Tesla P40·24 GB VRAM

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

Yes. The Tesla P40 runs cilla-mini fully in VRAM at FP16 / BF16 (about 0.6 GB of 24 GB).

best quant
FP16 / BF16
needs
0.6 GB
usable
24 GB

Estimates from the model's size.Try other contexts on the cilla-mini page →

Every quantisation of cilla-mini on the Tesla P40

QuantMemoryFit
Q4_K_M 0.5 GBfits
Q8_0 0.5 GBfits
FP16 / BF16 pick0.6 GBfits

Speed is a bandwidth-based estimate for generating text (347 GB/s); prompt processing and long contexts are slower.

TTS models that fit the Tesla P40 comfortably

All that fit →
ModelParamsBest quantMemoryLicence
facebook logomms-tts-engfacebook
36MFP16 / BF16≈0.6 GBFits in memoryNon-commercial
hexgrad logoKokoro-82Mhexgrad
82MFP16 / BF16≈0.7 GBFits in memoryCommercial OK
Supertone logosupertonic-2Supertone
66MFP16 / BF16≈0.7 GBFits in memoryConditions
Supertone logosupertonicSupertone
66MFP16 / BF16≈0.7 GBFits in memoryConditions
82MFP16 / BF16≈0.7 GBFits in memoryCommercial OK
OpenMOSS-Team logoMOSS-TTS-Nano-100MOpenMOSS-Team
100MFP16 / BF16≈0.7 GBFits in memoryCommercial OK

Other GPUs that run cilla-mini

Frequently asked questions

Can the Tesla P40 run cilla-mini?
Yes. The Tesla P40 runs cilla-mini fully in VRAM at FP16 / BF16 (about 0.6 GB of 24 GB).
Which quantisation of cilla-mini should I use on the Tesla P40?
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.