HHostingByteSoftware
Ctrl K
ComunidadGuías…

Ayuda y problemas de node-llama-cpp

Pregunta a la comunidad o invita a KiBot con contexto del repositorio.

Publicar un problema

No publiques contraseñas, tokens ni claves privadas. HostingByte aplica una redacción básica de secretos evidentes antes de guardar el texto.

Problemas en HostingByte

Todavía no hay problemas publicados.

Issues abiertas upstream

#683 bug: `getEmbeddingFor()` never prepends BOS for UGM-vocab models, even when the GGUF sets `add_bos_token = true` (breaks bge-m3 CLS pooling) · 1 comentarios#684 bug: a custom logger stops receiving llama.cpp log lines after a second getLlama() call in the same process · 0 comentarios#682 bug: node-llama-cpp installs 664 MB on linux/x64/glibc · 0 comentarios#580 Windows CUDA: cuMemAddressReserve failure in VMM pool causes hard abort (GGML_CUDA_NO_VMM workaround) · 2 comentarios#674 bug: Embeddings from non-causal models silently diverge when the input is longer than the context's batch size · 0 comentarios#590 Segmentation fault during embedBatch on ARM64 Linux (aarch64) · 3 comentarios#676 bug: Windows: llama-addon.node is unsigned and blocked by Smart App Control (error 4551), please consider code signing · 0 comentarios#672 bug: template chat wrapper silently accepts repeated placeholders · 0 comentarios#669 Dependency: bump simple-git past unpatched CVE range (pins ^3.36.0; only 4.0.2 is patched) · 0 comentarios