Jina-VLM: Small Multilingual Vision Language Model
AI & ML interests
Search foundation: embeddings, rerankers, small LMs for better search
Recent Activity
Papers
Jina-VLM: Small Multilingual Vision Language Model
jina-reranker-v3: Last but Not Late Interaction for Document Reranking
high quality code embeddings trained from code generation models
-
Efficient Code Embeddings from Code Generation Models
Paper • 2508.21290 • Published • 19 -
jinaai/jina-code-embeddings-1.5b
Feature Extraction • 2B • Updated • 4.37k • 30 -
jinaai/jina-code-embeddings-0.5b
Feature Extraction • 0.5B • Updated • 20.6k • 17 -
jinaai/jina-code-embeddings-1.5b-GGUF
2B • Updated • 2.16k • 13
max. ~1000 images and OCR text included
Multilingual multi-task general text embedding model
Multimodal text-image embeddings
-
jina-clip-v2: Multilingual Multimodal Embeddings for Text and Images
Paper • 2412.08802 • Published • 5 -
Jina CLIP: Your CLIP Model Is Also Your Text Retriever
Paper • 2405.20204 • Published • 37 -
jinaai/jina-clip-v2
Feature Extraction • 0.9B • Updated • 197k • 297 -
jinaai/jina-clip-v1
Feature Extraction • 0.2B • Updated • 80.9k • 256
The V2 family of Jina Embeddings supports encoding large documents with 8k sequence length.
-
Jina Embeddings 2: 8192-Token General-Purpose Text Embeddings for Long Documents
Paper • 2310.19923 • Published • 14 -
Multi-Task Contrastive Learning for 8192-Token Bilingual Text Embeddings
Paper • 2402.17016 • Published • 5 -
jinaai/jina-embeddings-v2-base-en
Feature Extraction • 0.1B • Updated • 134k • 731 -
jinaai/jina-embeddings-v2-base-zh
Feature Extraction • 0.2B • Updated • 255k • 245
Neural Reranker models for English language
0.6B Listwise Reranker for SOTA Multilingual Retrieval
Universal Embeddings for Multimodal Multilingual Retrieval
-
jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval
Paper • 2506.18902 • Published • 12 -
jinaai/jina-embeddings-v4
Visual Document Retrieval • 4B • Updated • 69.4k • 429 -
jinaai/jina-embeddings-v4-text-retrieval-GGUF
3B • Updated • 4.56k • 24 -
jinaai/jina-embeddings-v4-text-matching-GGUF
3B • Updated • 6.99k • 6
A copy of Jina VDR in BEIR format for usage with MTEB
Convert HTML content to LLM-friendly Markdown/JSON content
A collection of state-of-the-art multilingual neural rerankers
This collection list our ColBERT like late interaction retriever models
A novel set of high-performance sentence embedding models.
-
Jina Embeddings: A Novel Set of High-Performance Sentence Embedding Models
Paper • 2307.11224 • Published • 6 -
jinaai/jina-embedding-l-en-v1
Sentence Similarity • Updated • 239 • 25 -
jinaai/jina-embedding-b-en-v1
Sentence Similarity • Updated • 5.07k • 8 -
jinaai/jina-embedding-s-en-v1
Sentence Similarity • Updated • 2.34k • 26
Jina-VLM: Small Multilingual Vision Language Model
0.6B Listwise Reranker for SOTA Multilingual Retrieval
high quality code embeddings trained from code generation models
-
Efficient Code Embeddings from Code Generation Models
Paper • 2508.21290 • Published • 19 -
jinaai/jina-code-embeddings-1.5b
Feature Extraction • 2B • Updated • 4.37k • 30 -
jinaai/jina-code-embeddings-0.5b
Feature Extraction • 0.5B • Updated • 20.6k • 17 -
jinaai/jina-code-embeddings-1.5b-GGUF
2B • Updated • 2.16k • 13
Universal Embeddings for Multimodal Multilingual Retrieval
-
jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval
Paper • 2506.18902 • Published • 12 -
jinaai/jina-embeddings-v4
Visual Document Retrieval • 4B • Updated • 69.4k • 429 -
jinaai/jina-embeddings-v4-text-retrieval-GGUF
3B • Updated • 4.56k • 24 -
jinaai/jina-embeddings-v4-text-matching-GGUF
3B • Updated • 6.99k • 6
A copy of Jina VDR in BEIR format for usage with MTEB
max. ~1000 images and OCR text included
Convert HTML content to LLM-friendly Markdown/JSON content
Multilingual multi-task general text embedding model
A collection of state-of-the-art multilingual neural rerankers
Multimodal text-image embeddings
-
jina-clip-v2: Multilingual Multimodal Embeddings for Text and Images
Paper • 2412.08802 • Published • 5 -
Jina CLIP: Your CLIP Model Is Also Your Text Retriever
Paper • 2405.20204 • Published • 37 -
jinaai/jina-clip-v2
Feature Extraction • 0.9B • Updated • 197k • 297 -
jinaai/jina-clip-v1
Feature Extraction • 0.2B • Updated • 80.9k • 256
This collection list our ColBERT like late interaction retriever models
The V2 family of Jina Embeddings supports encoding large documents with 8k sequence length.
-
Jina Embeddings 2: 8192-Token General-Purpose Text Embeddings for Long Documents
Paper • 2310.19923 • Published • 14 -
Multi-Task Contrastive Learning for 8192-Token Bilingual Text Embeddings
Paper • 2402.17016 • Published • 5 -
jinaai/jina-embeddings-v2-base-en
Feature Extraction • 0.1B • Updated • 134k • 731 -
jinaai/jina-embeddings-v2-base-zh
Feature Extraction • 0.2B • Updated • 255k • 245
A novel set of high-performance sentence embedding models.
-
Jina Embeddings: A Novel Set of High-Performance Sentence Embedding Models
Paper • 2307.11224 • Published • 6 -
jinaai/jina-embedding-l-en-v1
Sentence Similarity • Updated • 239 • 25 -
jinaai/jina-embedding-b-en-v1
Sentence Similarity • Updated • 5.07k • 8 -
jinaai/jina-embedding-s-en-v1
Sentence Similarity • Updated • 2.34k • 26
Neural Reranker models for English language