From 813bcd290cad58ded4689617d7b3c55cacb23669 Mon Sep 17 00:00:00 2001
From: agent <agent@Studio.local>
Date: Thu, 17 Oct 2024 11:09:31 +0800
Subject: [PATCH] chore: add qwen2.5 32b awq

---
 src/recipe.yaml | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/src/recipe.yaml b/src/recipe.yaml
index 7b5ed356..6788baf9 100644
--- a/src/recipe.yaml
+++ b/src/recipe.yaml
@@ -976,3 +976,20 @@
       memory: 60Gi
     traffic:
       timeout: 300
+'qwen2.5:32b-instruct-awq-4bit':
+  project: vllm-chat
+  extra_labels:
+    model_name: Qwen/Qwen2.5-32B-Instruct-AWQ
+    openllm_alias: 32b-4bit
+    platforms: linux
+  engine_config:
+    max_model_len: 20480
+    model: Qwen/Qwen2.5-32B-Instruct-AWQ
+  project: vllm-chat
+  service_config:
+    name: qwen2.5
+    resources:
+      gpu: 1
+      gpu_type: nvidia-tesla-a100
+    traffic:
+      timeout: 300