michaelfeil / infinity

Infinity is a high-throughput, low-latency REST API for serving vector embeddings, supporting a wide range of text-embedding models and frameworks.

https://michaelfeil.eu/infinity/

MIT License

976 stars 72 forks source link

Closed michaelfeil closed 4 weeks ago

michaelfeil commented 3 months ago

Add support for MultiModal Inference / Clip

tjtanaa commented 3 months ago

How about adding API support to JinaAi's clip-as-a-service or Bentoml's clip-as-a-service ? JinaAi's clip-as-a-service also supports running clip models on onnx and tensorrt.

michaelfeil commented 1 month ago

@tjtanaa Both have slightly outdated tech / few recent commits. I think https://github.com/jina-ai/clip-as-service got Jina started in the early days

vlassisemm commented 1 month ago

Hey, any progress here?

michaelfeil commented 1 month ago

@vlassisemm Actually hacking in Aleksa's Discord server live on it https://discord.gg/dtNDQhPJ (back at 3pm PST)

michaelfeil commented 1 month ago

249 @vlassisemm