Tech2wild/GLM-5.3-Int4-Int8Mix

Card: weights live + public, benchmark table (46 tok/s c6), GitHub recipe link

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Card: verified evidence (structural + byte-for-byte QuantTrio match + coherence), honest no-parity caveat, sm12x overlay serving note

Tighten credits: Neural Magic (compressed-tensors/llm-compressor), QuantTrio (GLM recipe), IncoAI (DFlash2)

Stage model card (weights uploading after quant)

initial commit