Traceback (most recent call last)

WARNING: found GPU Tesla K80: limited to total length < 1000
Downloading alphafold2 weights to .: 100%|██████████| 3.47G/3.47G [00:39<00:00, 94.6MB/s]
2022-02-25 09:41:06,761 Running colabfold 1.2.0 (b66951a9c01b94c535e084682b413ce7fc3c2169)
2022-02-25 09:41:06,779 Found 8 citations for tools or databases

RuntimeError Traceback (most recent call last)
in ()
50 pair_mode=pair_mode,
51 stop_at_score=float(100),
---> 52 prediction_callback=prediction_callback,
53 )

10 frames
/usr/local/lib/python3.7/dist-packages/jax/interpreters/xla.py in _execute_compiled(name, compiled, output_buffer_counts, handlers, kept_var_idx, *args)
1098 for i, x in enumerate(args)
1099 if x is not token and i in kept_var_idx))
-> 1100 out_bufs = compiled.execute(input_bufs)
1101 check_special(name, out_bufs)
1102 if output_buffer_counts is None:

RuntimeError: RESOURCE_EXHAUSTED: Out of memory while trying to allocate 50331648 bytes.

Reproduce the failing notebook cell shown in the traceback with the reported Tesla K80 warning, then inspect the JAX execution that raises RESOURCE_EXHAUSTED. Determine whether the failure is specific to the reported input or GPU limits; done should be a verified fix or a clearly documented limitation.