Jesse Gross 172b5924af llm: Avoid integer underflow on llama engine memory layout ...

On the llama engine, when we compute the memory layout, we reserve
a buffer to allow for some flexibility for incorrect estimates.
This is subtracted from GPU free memory and on GPUs with limited
memory, it may underflow.

Fixes #13494

2025-12-19 15:48:15 -08:00

..

llm_darwin.go

Optimize container images for startup (#6547)

2024-09-12 12:10:30 -07:00

llm_linux.go

Optimize container images for startup (#6547)

2024-09-12 12:10:30 -07:00

llm_windows.go

win: lint fix (#10571)

2025-05-05 11:08:12 -07:00

server_test.go

llm: Don't always evict models on CPU-only systems

2025-12-02 10:58:08 -08:00

server.go

llm: Avoid integer underflow on llama engine memory layout

2025-12-19 15:48:15 -08:00

status.go

logs: catch rocm errors (#12888)

2025-10-31 09:54:25 -07:00