Model Capabilities - Voyage-context-3 支持 2048, 1024, 512 和 256 维度，并具备量化功能 [1] Cost Efficiency - Voyage-context-3 (int8, 2048 维度) 相比 OpenAI-v3-large (float, 3072 维度) 降低了 83% 的向量数据库成本 [1] Performance - Voyage-context-3 提供了 860% 更好的检索质量 [1]

Performance Summary - Voyage-context-3 outperforms all models across all domains in 93 retrieval datasets spanning nine domains [1] - Voyage-context-3 outperforms OpenAI-v3-large by 1420 basis points (14.2%) [1] - Voyage-context-3 outperforms Cohere-v4 by 789 basis points (7.89%) [1] - Voyage-context-3 outperforms Jina-v3 by 2366 basis points (23.66%) [1]

Technically, unlike traditional chunk embedding, the model processes the entire doc in a single pass to embed each chunk.This way, it sees all the chunks at the same time to generate global document-aware chunk embeddings.This gives semantically aware retrieval in RAG. https://t.co/LFIGCGwjtC ...

voyage-context-3 embedding model by @MongoDB solves this.It is a contextualized chunk embedding model that produces vectors for chunks that capture the full document context without any manual metadata and context augmentation.Check this visual 👇 https://t.co/3gJT2pese5 ...

Despite tuning and balancing tradeoffs, the final chunk embeddings are generated independently with no interaction with each other.This isn't true with real-world docs, which have long-range dependencies.Check this 👇 https://t.co/3pydkHDNCN ...

Chunking Challenges in RAG - Chunking involves determining overlap and generating summaries, which can be complex [1] - Lack of chunking increases token costs [1] - Large chunks may result in loss of fine-grained context [1] - Small chunks may result in loss of global/neighbourhood context [1]

RAG is 80% retrieval and 20% generation.So if RAG isn't working, most likely, it's a retrieval issue, which further originates from chunking and embedding.Contextualized chunk embedding models solve this.Let's dive in to learn more! https://t.co/vnQ5tAj1oe ...

Model Performance - A new embedding model outperforms OpenAI and Cohere models [1] - The new model reduces vector database costs by approximately 200 times [1]

RT Avi Chawla (@_avichawla)Generative vs. discriminative models in ML:(a popular ML interview question) https://t.co/bOWdx8CywA ...

Generative vs. discriminative models in ML:(a popular ML interview question) https://t.co/bOWdx8CywA ...