From 782440431942eea037cb80939caa7727418e7d12 Mon Sep 17 00:00:00 2001
From: Storme-bit <tk.stomre@gmail.com>
Date: Sun, 19 Apr 2026 07:59:31 -0700
Subject: [PATCH] fixed token count reading

---
 packages/inference-service/src/providers/llamacpp.js | 5 +++++
 packages/inference-service/src/routes/inference.js   | 1 +
 packages/orchestration-service/src/chat/index.js     | 1 +
 3 files changed, 7 insertions(+)

diff --git a/packages/inference-service/src/providers/llamacpp.js b/packages/inference-service/src/providers/llamacpp.js
index 69eb6ac..46db8e1 100644
--- a/packages/inference-service/src/providers/llamacpp.js
+++ b/packages/inference-service/src/providers/llamacpp.js
@@ -88,6 +88,11 @@ async function* completeStream(prompt, options = {}) {
         if (delta) yield { response: delta, done: false };
     }
   }
+
+  console.log('[llamacpp] finalTokenCount:', finalTokenCount);
+yield { response: '', done: true, model: finalModel, tokenCount: finalTokenCount };
+
+
   yield { response: '', done: true, model: finalModel, tokenCount: finalTokenCount };
 }
 
diff --git a/packages/inference-service/src/routes/inference.js b/packages/inference-service/src/routes/inference.js
index 4b23c52..d850750 100644
--- a/packages/inference-service/src/routes/inference.js
+++ b/packages/inference-service/src/routes/inference.js
@@ -42,6 +42,7 @@ router.post('/complete/stream', async (req, res) => {
                 // capture final metadata from the done signal
                 lastModel  = chunk.model      ?? lastModel;
                 tokenCount = chunk.tokenCount ?? tokenCount;
+                console.log('[inference router] tokenCount from chunk:', chunk.tokenCount, '→', tokenCount);
             }
         }
 
diff --git a/packages/orchestration-service/src/chat/index.js b/packages/orchestration-service/src/chat/index.js
index e525315..7707124 100644
--- a/packages/orchestration-service/src/chat/index.js
+++ b/packages/orchestration-service/src/chat/index.js
@@ -317,6 +317,7 @@ async function chatStream(externalId, userMessage, onChunk, options = {}) {
     console.log("[orchestration] final streamed text length:", fullText.length);
 
     if (fullText.trim()) {
+      console.log('[chat] tokenCount before save:', tokenCount);
       await memory.createEpisode(session.id, userMessage, fullText, tokenCount, session.project_id ?? null);
       const allEpisodes = await memory.getRecentEpisodes(session.id, 9999);
       triggerSummary(session, allEpisodes);