Packaging a GGUF model fails with 404 “model not found” when using docker model package

[ilopezluna]

docker model package --gguf $PWD/model.F16.gguf repo/name:tag
[...]
Transferred: 948.10 MB
Failed to package model
package model: failed to load packaged model: tag model: tagging failed with status 404 Not Found: tagging model: model not found

Despite the model transfer succeeding, the CLI fails when tagging the packaged model. This suggests either:

• the model is not correctly written to the registry before tagging
• the model is not properly retrieved after tagging

Impact:
Model packaging is broken for GGUF models, preventing users from publishing models to a registry.

Proposed action:

• Investigate the packaging flow for GGUF models.
• Add integration tests (see current integration tests)
• Fix the underlying cause of the 404 tagging failure