Mikhail Kim

The same for laion/CLIP-ViT-B-32-xlm-roberta-base-laion5B-s13B-b90k

if I try to load model without HF, this error encountered: ```python import open_clip model, _, preprocess = open_clip.create_model_and_transforms('xlm-roberta-large-ViT-H-14', pretrained='frozen_laion5b_s13b_b90k') Downloading (…)lve/main/config.json: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 616/616 [00:00

What about first error, when config can't be found?

> are you running on an x86? yes

As I know, ann-benchmarks works with specific dataset format (i.e h5df), so converting is required. By the way, [multimodal cc3m](https://huggingface.co/datasets/unum-cloud/ann-cc-3m) (bigg clip image and text embeddings) and [arxiv title-abstract](https://huggingface.co/datasets/unum-cloud/ann-arxiv-2m) (e5...