Systems and methods for compression of artificial intelligence

A method comprising:

• storing compressed data in a storage device at one or more address ranges;
• maintaining a mapping between the one or more address ranges and respective compression types;
• receiving a request for data associated with an address;
• identifying an address range corresponding to the address based on the mapping;
• and
• selecting a decompression circuit corresponding to the address range to decompress the compressed data.

A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause a system to:

• maintain a mapping between a set of address ranges of a storage device and respective compression types;
• receive a request for compressed data associated with an address;
• identify an address range corresponding to the address based on the mapping;
• and
• select a decompression operation corresponding to the address range to decompress the compressed data.