mirror of
https://github.com/ggerganov/llama.cpp
synced 2026-03-03 13:50:01 +01:00
* graph : avoid branches between embedding and token inputs * models : make deepstack graphs (e.g. Qwen3 VL) have constant topology * ci : enable -DGGML_SCHED_NO_REALLOC=ON for server CI * cont : pad token embeddings to n_embd_inp |
||
|---|---|---|
| .. | ||
| actions | ||
| ISSUE_TEMPLATE | ||
| workflows | ||
| labeler.yml | ||
| pull_request_template.md | ||