experiments to make it supported by TEI
still getting this error even after removing prefix new from safetensors
2024-08-17T05:35:52.669949Z INFO download_artifacts: text_embeddings_core::download: core/src/download.rs:32: Model artifacts downloaded in 16.645255649s
2024-08-17T05:35:52.925979Z WARN text_embeddings_router: router/src/lib.rs:195: Could not find a Sentence Transformers config
2024-08-17T05:35:52.925991Z INFO text_embeddings_router: router/src/lib.rs:199: Maximum number of tokens per request: 8192
2024-08-17T05:35:52.926198Z INFO text_embeddings_core::tokenization: core/src/tokenization.rs:28: Starting 8 tokenization workers
2024-08-17T05:35:53.919394Z INFO text_embeddings_router: router/src/lib.rs:241: Starting model backend
2024-08-17T05:35:54.035317Z INFO text_embeddings_backend_candle: backends/candle/src/lib.rs:354: Starting FlashGTE model on Cuda(CudaDevice(DeviceId(1)))
Error: Could not create backend
Caused by:
Could not start backend: classifier model type is not supported for GTE
(venv) sigridjineth@sigridjineth-Z590-VISION-G:~$
I think only bert and roberta are supported for rerankers
You could open an issue as a feature request to add gte as a classifier
Any update on this?