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 Tesla T4 run NLD-Uni?

NLD-Uni·18.3B·Tesla T4·16 GB VRAM

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

Yes. The Tesla T4 runs NLD-Uni fully in VRAM at Q5_K_M (about 14 GB of 16 GB), at roughly 14 tokens/s.

best quant
Q5_K_M
needs
14 GB
usable
16 GB
est. speed
~14 tok/s

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

Every quantisation of NLD-Uni on the Tesla T4

QuantQualityMemoryFitSpeed
Q2_K Heavy loss7.4 GBfits~29 tok/s
Q3_K_M Noticeable loss10 GBfits~20 tok/s
Q4_K_M Small loss12 GBfits~16 tok/s
Q5_K_M pickNear-lossless14 GBfits~14 tok/s
Q6_K Near-lossless16 GBoffload~11 tok/s
Q8_0 Lossless21 GBoffload~4.6 tok/s
FP16 / BF16 Lossless37 GBoffload~1.4 tok/s

Speed is a bandwidth-based estimate for generating text (320 GB/s); prompt processing and long contexts are slower.

Measured speed on the Tesla T4

No community reports for this pair yet.

Report your own speed →

Omni models that fit the Tesla T4 comfortably

All that fit →
ModelParamsBest quantMemorySpeedLicence
17.5BQ6_K≈9.3 GBFits in memory~12 tok/sCommercial OK
12BQ6_K≈13 GBFits in memory~18 tok/sCommercial OK
ByteDance-Seed logoBAGEL-7B-MoTByteDance-Seed
14.7BQ6_K≈13 GBFits in memory~15 tok/sCommercial OK
10.7BQ8_0≈8.9 GBFits in memory~15 tok/sCommercial OK
HiDream-ai logoHiDream-E1-FullHiDream-ai
17.1BQ5_K_M≈15 GBFits in memory~14 tok/sCommercial OK
12BQ6_K≈13 GBFits in memory~18 tok/sCommercial OK

Other GPUs that run NLD-Uni

Frequently asked questions

Can the Tesla T4 run NLD-Uni?
Yes. The Tesla T4 runs NLD-Uni fully in VRAM at Q5_K_M (about 14 GB of 16 GB), at roughly 14 tokens/s.
How fast does NLD-Uni run on the Tesla T4?
An estimate from memory bandwidth gives about 14 tokens/s at Q5_K_M.
Which quantisation of NLD-Uni should I use on the Tesla T4?
Q5_K_M: 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.