Spaces:

kadirnar
/

Open-Sora

Running on Zero

cbensimon HF staff commited on Mar 19

Commit

496445e

•

1 Parent(s): 7aec4dc

Reduce ZeroGPU requested duration

Generation took 107s on my side (requesting too much time will force users to (un-necessary) wait a lot for their quotas to recharge)

Files changed (1) hide show

app.py CHANGED Viewed

@@ -30,7 +30,7 @@ import glob
 subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
-@spaces.GPU(duration=500)
 def run_inference(prompt_text):
     repo_id = "hpcai-tech/Open-Sora"

 subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
+@spaces.GPU(duration=200)
 def run_inference(prompt_text):
     repo_id = "hpcai-tech/Open-Sora"