Why are PyTorch .bin weights ignored when provisioning the Llama 3 model? #2
Answered
by
AhmadHassan-BTed
AhmadHassan-BTed
asked this question in
Q&A
|
In |
Answered by
AhmadHassan-BTed
Jun 27, 2026
Replies: 1 comment
|
This is a deliberate optimization. The script is configured to explicitly target Safetensors ( Safetensors is a safer, faster-loading format that is fully supported by modern inference frameworks like Hugging Face Transformers, vLLM, and TGI. Excluding |
0 replies
Answer selected by
AhmadHassan-BTed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
This is a deliberate optimization. The script is configured to explicitly target Safetensors (
.safetensors) weights and skip PyTorch.bin/.pthfiles.Safetensors is a safer, faster-loading format that is fully supported by modern inference frameworks like Hugging Face Transformers, vLLM, and TGI. Excluding
.binprevents downloading duplicate weights and saves ~16GB of download bandwidth and disk space.