Huaiyi Zhao
Huaiyi Zhao
Any updates on this feature?
I found something at #61, where you can use `{{entry.data.fields.keywords}} ` to extract tags of Zotero and do some tricks like `{{#each entry.data.fields.keywords}}{{#if @last}}#{{this}}{{else}}#{{this}}, {{/if}}{{/each}}` to convert them to obsidian...
Zotero is fantastic, but not developing this feature seems to me more of a business decision rather than a technical issue considering their charge for storage :'(
Yes, I am using 8*H20 myself. But I lower the `actor.megatron.expert_model_parallel_size` to `4` and `actor.megatron.tensor_model_parallel_size` to `2` which I think is compatible with 8 GPUs.
I have built a docker image for it `docker pull zhyict/verl:verl0.7.0-transformers4.57.1-vllm0.11.0-mcore0.13.1`
nvidia-cudnn-cu12 9.10.2.21 nvidia-cudnn-frontend 1.10.0 transformer_engine 2.2.0+d0c452c
For aync vllm rollout, see https://github.com/volcengine/verl/pull/3934
Try to uninstall mbridge and reinstall from github `pip install -U git+https://github.com/ISEEKYAN/mbridge.git`
@DBMing You'd better use the latest main of verl
The error seems vllm related.