Clean Nexum container metadata paths

Clean public model card metadata wording

chore(hub): remove library_name download-count metadata

chore(hub): remove download-tracking sidecars

Update README.md

Release Nexum-Expanded causal generation runtime (#6)

Remove redundant per-profile context column (#5)

Expand Nexum tool-building and enterprise capability overview (#4)

Update README.md

Clarify profile downloads and model entrypoints

Add Hub download query files and library_name=nexum (README.md)

Add Hub download query files and library_name=nexum (meta.yaml)

Add Hub download query files and library_name=nexum (params.json)

Add Hub download query files and library_name=nexum (config.json)

Make 4,194,304-token context a larger hero claim

README intro: bold 4 million token context wording

Add industry-leading 4M context intro to README

Highlight Nexum self-built tools on capability gaps

Upload README.md with huggingface_hub

Upload README.md with huggingface_hub

Upload README.md with huggingface_hub

Strengthen full-scale release marketing CTA

Convert BFCLv4 section to compact stats strip

Update BFCLv4 section placement and formatting

Add open attribution license and full-scale contact path

Improve Nexum model card and launch guides

Add BFCL v4 comparison to model card

Document native context windows and Lite tool focus

Pin container tool CLIs to the supported Node runtime

Correct profile container commands and private distribution status

Correct centralized container layout documentation

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

Continue organized Nexum-Universal tensor upload

Continue organized Nexum-Lite tensor upload

Organize private Nexum release into Lite, Universal, and Expanded profiles