e8c3ed3d96
Replace prompt_logprobs-based scoring with the new vLLM /v1/score endpoint. Much simpler: one API call per memory drop, returns per-message total_logprob directly. No chunking needed, no OOM risk — the endpoint only computes logits for scored tokens. Co-Authored-By: Proof of Concept <poc@bcachefs.org> |
||
|---|---|---|
| .. | ||
| api | ||
| tools | ||
| cli.rs | ||
| dmn.rs | ||
| identity.rs | ||
| log.rs | ||
| mod.rs | ||
| observe.rs | ||
| parsing.rs | ||
| runner.rs | ||
| tui.rs | ||
| types.rs | ||
| ui_channel.rs | ||