From 16c66e968c08167e1891735fce39df9628a1ae94 Mon Sep 17 00:00:00 2001
From: hiro <vuonghoainam.work@gmail.com>
Date: Tue, 12 Dec 2023 05:58:08 +0700
Subject: [PATCH] chore: Update n_threads to cpu_threads

---
 core/src/types/index.ts                                     | 1 +
 extensions/inference-nitro-extension/src/@types/global.d.ts | 1 +
 extensions/inference-nitro-extension/src/index.ts           | 2 +-
 extensions/inference-nitro-extension/src/module.ts          | 2 +-
 extensions/inference-openai-extension/src/index.ts          | 1 -
 5 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/core/src/types/index.ts b/core/src/types/index.ts
index d5b51cfc0..81ea7e14e 100644
--- a/core/src/types/index.ts
+++ b/core/src/types/index.ts
@@ -275,6 +275,7 @@ export type ModelSettingParams = {
   ngl?: number;
   embedding?: boolean;
   n_parallel?: number;
+  cpu_threads: number;
   system_prompt?: string;
   user_prompt?: string;
   ai_prompt?: string;
diff --git a/extensions/inference-nitro-extension/src/@types/global.d.ts b/extensions/inference-nitro-extension/src/@types/global.d.ts
index f93a3e4c9..62eb65e52 100644
--- a/extensions/inference-nitro-extension/src/@types/global.d.ts
+++ b/extensions/inference-nitro-extension/src/@types/global.d.ts
@@ -12,6 +12,7 @@ declare const INFERENCE_URL: string;
 interface EngineSettings {
   ctx_len: number;
   ngl: number;
+  cpu_threads: number;
   cont_batching: boolean;
   embedding: boolean;
 }
diff --git a/extensions/inference-nitro-extension/src/index.ts b/extensions/inference-nitro-extension/src/index.ts
index e5f3f4360..f2fbf0d34 100644
--- a/extensions/inference-nitro-extension/src/index.ts
+++ b/extensions/inference-nitro-extension/src/index.ts
@@ -12,7 +12,6 @@ import {
   EventName,
   MessageRequest,
   MessageStatus,
-  ModelSettingParams,
   ExtensionType,
   ThreadContent,
   ThreadMessage,
@@ -41,6 +40,7 @@ export default class JanInferenceNitroExtension implements InferenceExtension {
   private static _engineSettings: EngineSettings = {
     ctx_len: 2048,
     ngl: 100,
+    cpu_threads: 1,
     cont_batching: false,
     embedding: false,
   };
diff --git a/extensions/inference-nitro-extension/src/module.ts b/extensions/inference-nitro-extension/src/module.ts
index b5ba69f5a..266566e91 100644
--- a/extensions/inference-nitro-extension/src/module.ts
+++ b/extensions/inference-nitro-extension/src/module.ts
@@ -53,7 +53,7 @@ async function initModel(wrapper: any): Promise<ModelOperationResponse> {
       llama_model_path: currentModelFile,
       ...wrapper.model.settings,
       // This is critical and requires real system information
-      n_threads: nitroResourceProbe.numCpuPhysicalCore,
+      cpu_threads: nitroResourceProbe.numCpuPhysicalCore,
     };
     log.info(`Load model settings: ${JSON.stringify(settings, null, 2)}`);
     return (
diff --git a/extensions/inference-openai-extension/src/index.ts b/extensions/inference-openai-extension/src/index.ts
index 7e3e6e71e..6bab563dd 100644
--- a/extensions/inference-openai-extension/src/index.ts
+++ b/extensions/inference-openai-extension/src/index.ts
@@ -12,7 +12,6 @@ import {
   EventName,
   MessageRequest,
   MessageStatus,
-  ModelSettingParams,
   ExtensionType,
   ThreadContent,
   ThreadMessage,