BUG: GTE-qwen2 Embedding Dimension error (#2565)

xorbitsai · Nov 21, 2024 · a49c1a6 · a49c1a6
1 parent 0cdfb43
commit a49c1a6
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 3 deletions.
diff --git a/doc/source/models/builtin/embedding/gte-qwen2.rst b/doc/source/models/builtin/embedding/gte-qwen2.rst
@@ -11,7 +11,7 @@ gte-Qwen2
 Specifications
 ^^^^^^^^^^^^^^
 
-- **Dimensions:** 4096
+- **Dimensions:** 3584
 - **Max Tokens:** 32000
 - **Model ID:** Alibaba-NLP/gte-Qwen2-7B-instruct
 - **Model Hubs**: `Hugging Face <https://huggingface.co/Alibaba-NLP/gte-Qwen2-7B-instruct>`__, `ModelScope <https://modelscope.cn/models/iic/gte_Qwen2-7B-instruct>`__

diff --git a/xinference/model/embedding/model_spec.json b/xinference/model/embedding/model_spec.json
@@ -233,7 +233,7 @@
   },
   {
     "model_name": "gte-Qwen2",
-    "dimensions": 4096,
+    "dimensions": 3584,
     "max_tokens": 32000,
     "language": ["zh", "en"],
     "model_id": "Alibaba-NLP/gte-Qwen2-7B-instruct",

diff --git a/xinference/model/embedding/model_spec_modelscope.json b/xinference/model/embedding/model_spec_modelscope.json
@@ -235,7 +235,7 @@
   },
   {
     "model_name": "gte-Qwen2",
-    "dimensions": 4096,
+    "dimensions": 3584,
     "max_tokens": 32000,
     "language": ["zh", "en"],
     "model_id": "iic/gte_Qwen2-7B-instruct",