Evandro Fonseca Junior d9198795bf fix(quick-1): bail on NULL KV cache and improve prefill error diagnostics ...

- engine.rs: new_session() now returns Err when llama_get_memory returns NULL
  instead of silently continuing, preventing decode corruption on 2nd request
- session.rs: decode_prefill error includes token count for easier diagnosis

2026-03-30 17:28:39 -03:00

..

api-contracts

Initial ForgeLM import

2026-03-29 19:10:57 -03:00

inference-core

fix(quick-1): bail on NULL KV cache and improve prefill error diagnostics

2026-03-30 17:28:39 -03:00

inference-server

Initial ForgeLM import

2026-03-29 19:10:57 -03:00

llama-sys

Initial ForgeLM import

2026-03-29 19:10:57 -03:00

localllm-cli

Initial ForgeLM import

2026-03-29 19:10:57 -03:00

metrics

Initial ForgeLM import

2026-03-29 19:10:57 -03:00