Hugging Face's logo

Qwen
/
Qwen3-VL-Reranker-2B

Text Ranking
Transformers
Safetensors
sentence-transformers
qwen3_vl
image-text-to-text
multimodal rerank
text rerank
Model card Files
xet
 Community
12
zyznull commited on
Commit
1a7345d
·
verified ·
1 Parent(s): 75a922c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -57,7 +57,7 @@ We utilize retrieval task datasets from various subtasks of [MMEB-v2](https://hu
57
 
58
  | Model | Size | MMEB-v2(Retrieval) - Avg | MMEB-v2(Retrieval) - Image | MMEB-v2(Retrieval) - Video | MMEB-v2(Retrieval) - VisDoc | MMTEB(Retrieval) | JinaVDR | ViDoRe(v3) |
59
  |-------|------|--------------------------|----------------------------|----------------------------|------------------------------|------------------|---------|------------|
60
- | Qwen3-VL-Embedding-2B | 2B | 73.6 | 74.9 | 52.1 | 80.2 | 68.1 | 71.0 | 52.9 |
61
  | jina-reranker-m0 | 2B | - | 68.2 | - | 85.2 | - | 82.2 | 57.8 |
62
  | Qwen3-VL-Reranker-2B | 2B | 75.1 | 73.8 | 52.1 | 83.4 | 70.0 | 80.9 | 60.8 |
63
  | Qwen3-VL-Reranker-8B | 8B | 79.2 | 80.7 | 55.8 | 86.3 | 74.9 | 83.6 | 66.7 |
@@ -87,7 +87,7 @@ model = Qwen3VLReranker(model_name_or_path=model_name_or_path)
87
  # Combine queries and documents into a single input list
88
 
89
  inputs = {
90
- "instruction": "Retrieval relevant image or text with user's query",
91
  "query": {"text": "A woman playing with her dog on a beach at sunset."},
92
  "documents": [
93
  {"text": "A woman shares a joyful moment with her golden retriever on a sun-drenched beach at sunset, as the dog offers its paw in a heartwarming display of companionship and trust."},
@@ -99,7 +99,7 @@ inputs = {
99
 
100
  scores = model.process(inputs)
101
  print(scores)
102
- # [0.8408790826797485, 0.6197134852409363, 0.7778129577636719]
103
  ```
104
  For more usage examples, please visit our [GitHub repository](https://github.com/QwenLM/Qwen3-VL-Embedding).
105
 
 
57
 
58
  | Model | Size | MMEB-v2(Retrieval) - Avg | MMEB-v2(Retrieval) - Image | MMEB-v2(Retrieval) - Video | MMEB-v2(Retrieval) - VisDoc | MMTEB(Retrieval) | JinaVDR | ViDoRe(v3) |
59
  |-------|------|--------------------------|----------------------------|----------------------------|------------------------------|------------------|---------|------------|
60
+ | Qwen3-VL-Embedding-2B | 2B | 73.4 | 74.8 | 53.6 | 79.2 | 68.1 | 71.0 | 52.9 |
61
  | jina-reranker-m0 | 2B | - | 68.2 | - | 85.2 | - | 82.2 | 57.8 |
62
  | Qwen3-VL-Reranker-2B | 2B | 75.1 | 73.8 | 52.1 | 83.4 | 70.0 | 80.9 | 60.8 |
63
  | Qwen3-VL-Reranker-8B | 8B | 79.2 | 80.7 | 55.8 | 86.3 | 74.9 | 83.6 | 66.7 |
 
87
  # Combine queries and documents into a single input list
88
 
89
  inputs = {
90
+ "instruction": "Retrieve images or text relevant to the user's query.",
91
  "query": {"text": "A woman playing with her dog on a beach at sunset."},
92
  "documents": [
93
  {"text": "A woman shares a joyful moment with her golden retriever on a sun-drenched beach at sunset, as the dog offers its paw in a heartwarming display of companionship and trust."},
 
99
 
100
  scores = model.process(inputs)
101
  print(scores)
102
+ # [0.8613124489784241, 0.6757137179374695, 0.8125371336936951]
103
  ```
104
  For more usage examples, please visit our [GitHub repository](https://github.com/QwenLM/Qwen3-VL-Embedding).
105