Can I run on a Tesla T4

[realsama]

I am trying to use google colab with a Tesla T4 attached but I haven't be able to successfully generate any images. The below is my stacktrace.

Global seed set to 27
Loading model from models/ldm/stable-diffusion-v1/model.ckpt
Global Step: 440000
UNet: Running in eps-prediction mode
Selected timesteps for ddim sampler: [  1  21  41  61  81 101 121 141 161 181 201 221 241 261 281 301 321 341
 361 381 401 421 441 461 481 501 521 541 561 581 601 621 641 661 681 701
 721 741 761 781 801 821 841 861 881 901 921 941 961 981]
Selected alphas for ddim sampler: a_t: tensor([0.9983, 0.9804, 0.9609, 0.9398, 0.9171, 0.8930, 0.8674, 0.8404, 0.8121,
        0.7827, 0.7521, 0.7207, 0.6885, 0.6557, 0.6224, 0.5888, 0.5551, 0.5215,
        0.4882, 0.4552, 0.4229, 0.3913, 0.3605, 0.3308, 0.3023, 0.2750, 0.2490,
        0.2245, 0.2014, 0.1799, 0.1598, 0.1413, 0.1243, 0.1087, 0.0946, 0.0819,
        0.0705, 0.0604, 0.0514, 0.0435, 0.0365, 0.0305, 0.0254, 0.0210, 0.0172,
        0.0140, 0.0113, 0.0091, 0.0073, 0.0058]); a_(t-1): [0.99914998 0.99829602 0.98038077 0.96087277 0.93978298 0.91713792
 0.89298052 0.86737001 0.84038192 0.81210774 0.78265446 0.75214338
 0.72070938 0.68849909 0.65566933 0.62238538 0.58881873 0.55514455
 0.52153981 0.4881804  0.45523876 0.42288151 0.39126703 0.36054322
 0.33084565 0.30229566 0.27499905 0.24904492 0.22450483 0.20143245
 0.1798636  0.15981644 0.14129217 0.12427604 0.10873855 0.09463691
 0.08191671 0.0705137  0.06035557 0.05136392 0.043456   0.03654652
 0.03054927 0.02537862 0.02095082 0.01718517 0.0140049  0.01133791
 0.00911731 0.00728173]
For the chosen value of eta, which is 0.0, this results in the following sigma_t schedule for ddim sampler tensor([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
        0., 0.], dtype=torch.float64)
^C

[gsajko]

Tesla T4 is enough in my opinion. But on "normal" colab we have to little of RAM - your output means exactly that.

[realsama]

Tesla T4 is enough in my opinion. But on "normal" colab we have to little of RAM - your output means exactly that.

I have been able to run it on an RTX 3090. Thank you.