Discussions

Good morning,

I want to verify whether my input token size exceeds the model’s limit. Is there a built-in function to calculate token length without making an API call and incurring costs? If using tiktoken is the only option, which encoding should I use?

You previously mentioned healthcare-specific embedding models. Any plans on making one available soon? Thank you!

in my test, when i send 60 ~ 70 requests in 5 seconds, the response time will gradually change from 1.5s to more than 20 seconds. I would like to ask whether there is a situation where the processing rate is degraded.

Hello, I would like to know if any of Voyage emedding models support Swedish and Finnish languages. If there are some comparisons with cohere multilingual - would be really interesting to know the results!

I'm using LangChain's QdrantVectorStore.from_documents(documents, embeddings,..) where embeddings is a VoyageAIEmbeddings("voyage-law-2",....). In the middle of uploading my documents, I got the error above. I've gotten this twice already. How can I fix this?

Hi. I have a few million documents that I generated embeddings for. Based on ad hoc experiments, it seems like for my purpose, a lower embedding dimensionality would be fine. I want to take my existing float/2048 dimension embeddings and reduce to int/1024. Obviously I can do whatever I want with my local embeddings, but is there a means to exactly replicate the quantization/dimension reduction done with the API?

Hi

Hi, do you have an SDK for Javascript/TypeScript?

We would like to integrate to our product. does it support hindi language?