server: Adding spec-decode counters to /metrics endpoint (#26389)

* * server: add spec-decode counters to /metrics endpoint * server: fixed review comments and now aligned param names exactly with vLLM.
2026-08-05 12:36:01 +02:00
parent 020760adfc
commit a035a88878
4 changed files with 58 additions and 0 deletions
@@ -532,6 +532,11 @@ struct server_task_result_metrics : server_task_result {
    uint64_t n_decode_total     = 0;
    uint64_t n_busy_slots_total = 0;

+    uint64_t n_draft_tokens_total      = 0;
+    uint64_t n_draft_accepted_total    = 0;
+    uint64_t n_draft_verif_steps_total = 0;
+    std::vector<uint64_t> n_accepted_per_pos_total;
+
    // while we can also use std::vector<server_slot> this requires copying the slot object which can be quite messy
    // therefore, we use json to temporarily store the slot.to_json() result
    json slots_data = json::array();