Update constant.ts

to add support for quantized(q4f16) qwen2-0.5b
mlc-ai · Jun 26, 2024 · f00cf05 · f00cf05
1 parent 6a06f4b
commit f00cf05
Showing 1 changed file with 17 additions and 0 deletions.
diff --git a/app/constant.ts b/app/constant.ts
@@ -347,6 +347,23 @@ export const DEFAULT_MODELS: ModelRecord[] = [
       temperature: 0.7,
       top_p: 0.95,
     },
+  },
+   {
+    name: "Qwen2-0.5B-Instruct-q4f16-MLC",
+    display_name: "Qwen",
+    provider: "Alibaba",
+    size: "0.5B",
+    quantization: "q4f16_1",
+    context_length: "4k",
+    family: "Qwen 2",
+    vram_required_MB: 500, //rough estimate
+    low_resource_required: true,
+    recommended_config: {
+      temperature: 0.7,
+      presence_penalty: 0,
+      frequency_penalty: 0,
+      top_p: 0.8,
+    },
   },
   {
     name: "Qwen2-0.5B-Instruct-q0f16-MLC",